Ultrasound-Targeted Nucleic Acid Delivery for Solid Tumor Therapy

Abstract

Depending upon multiple factors, malignant solid tumors are conventionally treated by some combination of surgical resection, radiation, chemotherapy, and immunotherapy. Despite decades of research, therapeutic responses remain poor for many cancer indications. Further, many current therapies in our armamentarium are either invasive or accompanied by toxic side effects. In lieu of traditional pharmaceutics and invasive therapeutic interventions, gene therapies offer more flexible and potentially more durable approaches for new anti-cancer therapies. Nonetheless, many current gene delivery approaches suffer from low transfection efficiency due to physiological barriers limiting extravasation and uptake of genetic material. Additionally, systemically administered gene therapies may lack target-specificity, which can lead to off-target effects. To overcome these challenges, many preclinical studies have shown the utility of focused ultrasound (FUS) to increase macromolecule uptake in cells and tissue under image guidance, demonstrating promise for improved delivery of therapeutics to solid tumors. As FUS-based drug delivery is now being tested in several clinical trials around the world, FUS-targeted gene therapy for solid tumor therapy may not be far behind. In this review, we comprehensively cover the literature pertaining to preclinical attempts to more efficiently deliver therapeutic genetic material with FUS and offer perspectives for future studies and clinical translation.

Graphical abstract

Introduction

As the standard of care for solid tumors has widened from resection, radiotherapy, and chemotherapy to include small molecule inhibitors and immunotherapies, more patients experience cancer control and remission today than even a decade ago. However, for many cancer indications, standard-of-care therapies do not achieve curative treatment results. Complications such as off-target effects, selective genetic pressure, and poor immune cell infiltration arise in patients refractory to more traditional pharmaceutical therapies [1]. These caveats may be circumvented by gene therapies, which aim to introduce genetic material into select cells or tissues, transforming cells by either up- or down-regulating gene expression [1]. In this review, we focus our attention on gene delivery approaches applied to solid tumors.

Nucleic acid delivery, leading to therapeutic modulation of gene expression, may be achieved using several approaches. These approaches have been expertly reviewed in detail by others before [2,3], thus we only introduce them here. Briefly, viral delivery systems present foreign genetic material by introducing replication-deficient viruses to target cells, which deliver nascent DNA by transduction [2]. Nevertheless, viral vectors can be immunogenic and may randomly insert portions of DNA into the host cell genome, causing potentially oncogenic mutagenesis [2]. Consequently, some have shied away from viral delivery methods, leading to the development of non-viral delivery systems such as short interfering RNA (siRNA), short hairpin RNA (shRNA), microRNA (miRNA), and plasmid delivery. siRNA and shRNA are two different classes of RNA molecules used for RNA interference (RNAi) or gene silencing applications [2,3]. Once in cells, both siRNA and shRNA result in mRNA inactivation or degradation of mRNA targets, silencing target protein production [2]. miRNAs also modulate gene expression through a variety of mechanisms, serving as potential therapeutic targets [2]. Plasmids are internalized by endocytosis, undergo endosomal escape, and are transported to the cell nucleus, where they carry out their role [2]. Lastly, Clustered Regularly-Interspaced Short Palindromic Repeats (CRISPR) technology has been developed for gene editing [4]. CRISPR systems are often delivered as plasmids which must undergo translation, yielding two critical components that complex together: (1) guide RNAs (gRNAs) which recognize the target sequence and (2) Cas9 protein which cuts the target region of the host genome [4]. This system allows for precise editing of an organism’s genome by cutting, deleting and/or inserting genes [4].

Such nucleic acid therapeutics may encounter several barriers between their administration and cellular internalization [2,5]. Upon systemic administration, circulating nucleic acids may be degraded or bind to blood serum proteins, hindering transport from blood vessels into surrounding tissue [5]. These limitations have motivated studies to explore protection of nucleic acids to prolong their half-lives and treatment windows. This includes encapsulation of the genetic material in liposomes or nanoparticles (NPs) [1,6] or complexing therapeutics to cationic polymers by electrostatic interactions [5]. Poor targeting of gene therapy to solid tumors can also hinder efficacy. Upon systemic administration, some genetic material may accumulate in tumors due to leaky tumor vasculature and decreased lymphatic drainage, a phenomenon called the enhanced permeability and retention (EPR) effect [7,8]. Therapeutic agents accumulate in tumors via the EPR effect through a variety of passive and active mechanisms: passage through inter-endothelial gaps, transcellular channels, or uptake through transcytosis [8]. However, delivery via the EPR effect typically lacks efficiency [7]. Indeed, relying on the EPR effect to deliver therapeutics may not afford sufficient agent accumulation, highlighting the need to explore more efficient avenues of delivery.

This review examines the application of an imaging technology (i.e. diagnostic ultrasound) reimagined into a therapeutic gene delivery modality. In an imaging context, ultrasound waves reflect off internal structures, creating echoes that are used to visualize patient anatomy [9]. Focusing the ultrasound energy (i.e. focused ultrasound; FUS) directs the acoustic waves to a specific target, like a solid tumor. In this manner, the acoustic energy can be deposited to either destroy tissue or enhance cell permeability with millimeter precision [9]. While various FUS modalities can be used for a range of therapeutic applications, including thermal tissue ablation and histotripsy, both of which deposit high amounts of energy in tissue [9], this review will focus on using FUS to enable gene delivery. Acoustically responsive 2–4 μm-sized gas-filled microbubbles (MBs), often made of a lipid or protein shell filled with a high molecular weight gas, are commonly administered intravenously in tandem with FUS. Many MBs, such as Optison^™, Definity®, and SonoVue®, are FDA-approved ultrasound contrast agents that are used in many of the FUS-mediated delivery applications cited in this review [10]. MBs expand, contract, and oscillate in response to acoustic pressure, amplifying the effects of FUS to transiently permeabilize vasculature and facilitate agent delivery [11]. FUS has thus emerged as a promising therapeutic modality to improve gene delivery.

In the past decade and a half, FUS has been employed in preclinical cancer models to increase genetic material uptake in tumors by permeabilizing cells [12], decreasing interstitial fluid pressure [13], modulating immune responses [14], and/or through unelucidated mechanisms. It has been proposed that mechanical stresses induced by MB-enhanced FUS (MB-FUS) permeabilizing vessels by disrupting tight junctions [15], permeabilizing cell membranes, and/or increasing endocytic activity [16]. Notably, for some preclinical cancer indications, FUS may be used in conjunction with magnetic resonance imaging (MRI) to more precisely visualize and guide treatment [17]. Overall, it is clear that the combination of FUS, MBs, and nucleotide-based therapeutic approaches have resulted in increased transfection efficiency correlated with tumor growth control in preclinical cancer models. In this review, we will cover FUS-mediated gene therapy applied to a wide breadth of solid tumor models, including glioma, hepatocellular carcinoma, colorectal cancer, pancreatic cancer, hormone-related cancers, skin cancers, and lung cancer. We acknowledge that this list is not comprehensive, and we note that our review focuses on studies with a significant in vivo component. Therefore, innovative approaches that have been advanced primarily in vitro may not be covered.

Glioma:

Glioma (GM) is the most common malignant primary brain tumor, accounting for 32% of central nervous system (CNS) tumors and 80% of malignant primary CNS tumors [18]. Current treatments, including surgical resection, chemotherapy, and radiotherapy, result in a median survival time of only 14.6 months [18]. Aside from poor prognosis, these treatments are often accompanied by detrimental side effects including permanent brain damage and chemotherapeutic toxicity [6,18–21]. GM’s high incidence rate, lack of non-invasive treatment options, and poor survival outcomes have motivated the development of new, clinically relevant treatments. While the successful development of such treatments for GM represents a tremendous challenge, MB-FUS has emerged as an appealing therapy to overcome these challenges.

A major impediment to delivering therapeutics to GM is the blood-brain barrier (BBB). The BBB protects the brain from circulating toxins and pathogens by controlling influx and efflux of molecules between the blood and the CNS [6]. Consequently, the BBB selectively inhibits delivery of systemically administered therapeutics into the brain, including in the infiltrating rim of GM, wherein cancer cells are protected by the BBB. Meanwhile, in the GM core, tumorigenesis results in leaky vasculature and heterogeneous flow [i.e. the blood-tumor barrier (BTB)]. In turn, the BTB also results in insufficient therapeutic accumulation in tumors [20]. Importantly, FUS-induced oscillation of MBs in the vasculature transiently disrupts both the BBB and BTB, allowing for non-invasive permeabilization of these barriers [6,19,20,22] (Fig. 1), modulation of interstitial flow [13], and thus improved therapeutic delivery.

Figure 1:

Mechanisms of FUS-induced permeation of the BBB in response to MB oscillation. The BBB can be permeabilized through three different bioeffects generated by oscillating MBs: disruption of tight junctions, sonoporation of the vascular endothelial cells and upregulation of transcytosis [22].

It has been proposed that the therapeutic efficacy of FUS-mediated treatment can be further improved by using nano- rather than micro-sized bubbles and carriers which may permeate through vasculature and accumulate in tumor interstitial space [20]. Yin et al. (2013) applied NB-FUS in a subcutaneous GM mouse model to silence SIRT2, an overexpressed anti-apoptotic gene, which resulted in a significant reduction of tumor volume and increased survival time. More recently, Guo et al. (2021) used MB-FUS to administer lipid-polymer hybrid (LPH) Smoothened- (SMO-) siRNA loaded NPs targeting the sonic hedgehog (SHH) subgroup in another primary brain tumor (i.e. medulloblastoma) [21]. Although different from glioma, medulloblastoma is the most common and malignant form of brain tumors in children and is characterized by activation of the SHH pathway [21]. Therefore, silencing SHH pathway promoters, such as SMO, serves as a potential therapeutic strategy. MB-FUS permeabilization of surrounding vasculature permitted NPs to overcome vascular and interstitial barriers, as indicated by a 10-fold increase in intracellular accumulation (Fig 2E, G) and a 5-fold increase in tissue penetration (Fig. 2B, D) in orthotopic tumors [21]. The enhanced accumulation (Fig. 2E) promoted SMO gene silencing and increased tumor cell apoptosis up to 16-fold [21]. To better understand LPH-siRNA NP transport dynamics and simulate time-consuming in vivo experiments, Guo et al (2021). used pharmacokinetic (PBPK) modeling of the TME network to model delivery of therapeutic agents. PBPK modeling identified design limitations and optimal NP and MB-FUS experimental parameters for the construct. Similar PBPK data can be used for further downstream analyses to refine delivery of MB-FUS siRNA and other therapies to brain tumors.

Figure 2: Improved LPH-siRNA extravasation, penetration, and cellular uptake in the GL261 glioma mouse tumors using MB-FUS.

A) Schematic demonstrating delivery of non-targeting LPH-siRNA. B) Representative fluorescent microscopy images of siRNA extravasation and penetration into glioma xenografts 8 hours post-treatment. C) Quantification of LPH in tumors without and without FUS 8 hours post treatment. D) Quantification of siRNA in tumors without and without FUS 8 hours post treatment. E) Quantification of siRNA delivery to cancer cells without and without FUS 8 hours post-treatment. F) Quantification of the ratio of LPH or siRNA uptake by cancer cells versus total cell uptake 8 hours post-treatment. G) Representative fluorescent microscopy images of LPH-siRNA uptake by cancer cells 8 hours post-LPH-siRNA administration. Green arrows indicate LPH-siRNA uptake by cancer cells and white arrows indicate LPH-siRNA uptake by brain cells. Plots show means ± SEM (N = 3). P values were determined by unpaired t tests. *P ≤ 0.05 and **P ≤ 0.01 [21].

Therapeutic delivery to tumors can be further enhanced by altering surface properties of gene delivery vehicles to target specific cell receptors [6,19]. In this section, we look at surface receptors associated with GM and their therapeutic potential. CD13, which is implicated in tumor progression, is a candidate for targeting as it is overexpressed by malignant tumors [6]. Zhao et al. (2018) demonstrated that CD13 targeted therapeutic delivery resulted in a 10-fold decrease in GM volume as compared to non-targeted therapy [6]. It was also noted that GM shrinkage rate decreased over time, likely due to metabolization of the therapeutic, suggesting that consecutive and recurring doses may improve results. Another candidate for targeted therapeutic delivery is vascular endothelial growth factor receptor 2 (VEGFR2), which is overexpressed by endothelial cells in many tumors [19] and critical for tumor angiogenesis. Chang et al. (2017) exploited this overexpression to target brain cancer cells by conjugating anti-VEGFR2 antibody to cationic MBs encapsulating a double suicide gene system: non-viral herpes simplex virus thymidine kinase/ganciclovir plasmid (pHSV-TK/GCV). The construct, which has shown promise in GM treatment, contains an enzyme-encoding plasmid that converts non-toxic prodrugs into toxic products, inhibiting DNA synthesis and killing tumor cells [19]. It was employed using FUS in orthotopic glioblastoma multiforme (GBM) tumor mouse models. GBM is the most malignant, heterogeneous, infiltrative, and unfortunately the most common, form of GM in adults [19]. This study demonstrated progress in GBM treatment development, as targeted MBs resulted in significantly increased gene transfection in tumors as compared to both non-targeted MBs and intratumoral injection. Survival and tumor growth inhibition were also improved by targeted MBs establishing a non-viral, non-invasive, and powerful GBM therapeutic.

Leaky vasculature in the TME contributes to high interstitial fluid pressure which alters and limits convective transport of systemically administered therapeutic agents [13]. Using brain penetrating NPs (BPNPs), Curley et al. (2020) demonstrated the potential of MB-FUS to augment interstitial fluid transport and improve therapeutic penetration in mouse GBM xenografts (Fig. 3). It was observed that MB-FUS increased interstitial flow velocity by two-fold and significantly altered direction of interstitial flow, resulting in increased BPNP tissue penetration [13]. Although this study focused on delivery of BPNPs, this augmentation of brain tumor interstitial flow may extend to a variety of therapeutic agents, including nucleic acids, to modulate tissue penetration.

Figure 3. Blood-tumor barrier opening with MRI–guided FUS markedly alters interstitial flow velocity in U87 gliomas.

(A) Pre- and post-FUS (0.55 MPa) T1-weighted contrast MRI sequences used for interstitial flow analyses. Arrow denotes enhancing tumor region used as the ROI for subsequent transport analysis. Gd, gadobenate dimeglumine contrast agent (MultiHance) administration. (B) Flow velocity magnitude map derived from the MRIs in (A). (C) Distribution of pre- and posttreatment voxel flow velocity magnitudes from (B). Red lines denote medians. (D and E) Plots of median flow velocity magnitudes (D) and Péclet numbers (E), pre- and posttreatment, with 0.45- and 0.55-MPa FUS. Paired data points are denoted by common colors and shapes. Bars, SEM. *P = 0.01, **P = 0.04, +P = 0.01, and ++P = 0.005 versus “Pre” at same PNP. Significance was assessed by two-way RM ANOVA followed by Sidak’s multiple comparisons tests. (F and G) Velocity direction changes in individual tumors due to BTB opening with 0.45-MPa (E) and 0.55-MPa (F) FUS. Each data point represents one voxel. (H) Mean velocity direction changes. Significance was tested by unpaired t test. n.s., not significant. Adapted from Curley et al. (2020) [13].

Hepatocellular Carcinoma:

Hepatocellular carcinoma (HCC) is the third most common cause of cancer-related deaths worldwide, with liver resection and transplantation as the most viable treatment options [23]. Success of these treatments is limited by the lack of organ donors, high recurrence, and late-stage diagnosis [23–27]. Furthermore, many candidates are denied treatment due to cirrhosis following HCC development [23–25]. Thus, prognosis of HCC patients is poor, and treatment typically results in only 2–3 months of extended survival as compared to untreated patients [24]. This inadequacy demonstrates the necessity to develop new, effective therapeutics for HCC. Previous studies have identified aberrant oncogenic gene expression in HCC tumors [23–29]. Consequently, abnormal gene expression in the TME has prompted FUS-mediated gene delivery to emerge as a potential therapeutic approach for HCC. In this section, we will review the three modes of gene delivery currently being explored for HCC treatment: (1) supplementation of tumor suppressors, (2) inhibition of anti-apoptotic genes, and (3) combinatory treatment.

Supplementation of tumor suppressors

miRNAs are non-coding small RNAs that regulate gene expression [23]. Abnormal expression of certain miRNAs in cancer, including in HCC, is instrumental in the migration and survival of tumor cells [23]. Thus, modulating miRNA expression in cancerous tissue is a pragmatic therapeutic approach. Specifically, studies have identified decreased levels of miR-122 in HCC, which regulates cell migration, cell proliferation, and chemoresistance [25]. Guo et al. (2020) recently treated subcutaneous HCC xenografts with US-aided miR-122 mimic nanodroplets (NDs) every three days for a total of five treatment cycles [25]. NDs, which can be used instead of MBs, are administered as a liquid and become MBs by acoustic droplet vaporization once triggered by US emissions [25]. Increased miR-122 expression and tumor volume reduction were identified in mice treated with miR-122 mimic NDs in comparison to groups treated without US and/or without miR-122 delivery [25]. However, miR-122 delivered with a commercially available jetPEI transfection reagent resulted in the most tumor growth suppression, demonstrating the need to further increase gene payload and improve transfection. In summary, this study provides a foundation for further refinement of miR-122 delivery as a promising therapeutic modality for HCC treatment.

Inhibition of anti-apoptotic genes

Oncogene expression also plays an important role in enabling liver cancer proliferation and survival [26,28,29]. Thus, many research groups have also explored the possibility of silencing such genes with FUS-mediated gene editing. In particular, class IA phosphoinositide 3-kinase (PI3K) plays a key role in hepatocarcinogenesis by recruiting enzymes to activate signaling pathways regulating cell growth, survival, proliferation, and motility [28]. Dong et al. (2020) investigated class IA PI3K activity inhibition by targeting one of its subunit genes: PIK3 CA. Furthermore, PIK3 CA itself is known to play a role in cell hyperplasia [28]. The delivery of four candidate PIK3-modulating miRNAs was performed in vivo: miRNA-139, −203a, −378a and −422. Plasmids encoding the miRNAs were loaded in nanodroplets and evaluated individually in subcutaneous HCC xenografts. Pre-miRNA-139 and −378a plasmids inhibited tumor growth up to 4 times as much as the control group and increased survival by up to 10 days [28], demonstrating the efficacy of restoring aberrant miRNA expression for tumor suppression.

Oncogene silencing has been similarly accomplished using small interfering RNA (siRNA). Abnormal gene expression, particularly upregulated anti-apoptotic genes, promotes drug resistance, which diminishes chemotherapeutic efficacy [23,24,26]. B-cell-lymphoma-2 (BCL-2) is an anti-apoptotic protein that reduces HCC cell apoptosis by suppressing cytochrome c and apoptosis-inducing factor (AIF) [26]. Yin et al. (2014) postulated that downregulation of BCL-2 would reverse acquired drug resistance, increase HCC cell apoptosis, and augment drug efficacy. They constructed NBs loaded with BCL-2-targeting siRNA and an anti-cancer drug, paclitaxel (PTX), and administered these at varying dosages to subcutaneous HCC in mice using FUS. Following treatment, they observed increased expression of pro-apoptotic Bax, indicating the regimen’s cytotoxicity to HCC cells. The therapeutic benefit of this treatment was further established by the significant increase in survival time, tumor shrinkage, and decrease in IC₅₀ [26]. Thus, this study demonstrated that codelivery of siRNA with a chemotherapeutic reestablished drug sensitivity and permitted lower drug dosage, potentially diminishing common chemotherapy cytotoxic effects.

As mentioned in the glioma section, spatial targeting of a therapeutic can be improved by modifying carrier surfaces with targeting ligands. Wu. et al leveraged the expression of glypican-3 (GPC3) which is unique to HCC cells to formulate targeted nanobubbles (TNB) conjugated with GPC3 antibody and neuroepithelial transforming gene 1 (NET-1) siRNA. NET-1 is responsible for regulating cell proliferation and is frequently overexpressed in HCC [29]. Administration of this construct in HCC subcutaneous mouse models resulted in increased survival time and decreased tumor volume in comparison to non-targeted nanobubbles [29]. Non-targeted controls showed scattered fluorescence along the abdominal aorta, whereas TNBs illustrated augmented accumulation of NET-1 siRNA in tumor tissue, providing more siRNA to silence NET-1 gene production in tumors. Overall, this demonstrates the ability TNB has to boost treatment efficacy.

Combinatory HCC therapies

A major challenge in cancer treatment is the multifaceted interplay of gene expression and cell signaling that impact cancer proliferation [23,24,27]. Silencing, knocking down, or increasing expression of a single factor may lack adequate power to control tumor growth and increase animal survival. Thus, multi-targeted therapeutics may offer superior performance compared to targeted monotherapies. Yu et al. (2013) employed a HSVtk/GCV double suicide gene system in tandem with tissue inhibitor of metalloproteinase 3 (TIMP3), a tumor suppressor, in HCC xenografts. The double suicide gene was driven by the human α-fetoprotein (AFP) tissue-specific promoter to selectively target apoptosis in HCC cells [27]. Furthermore, TIMP3 delivery increased low endogenous TIMP3 levels in HCC tumors [30]. In vitro analysis demonstrated that the greatest amount of apoptosis and angiogenesis inhibition was achieved with synergistic therapies as opposed to TIMP3 or HSVtk/GCV alone [27]. This was further corroborated by in vivo experiments yielding the lowest tumor volume and the highest percent survival in the combinatorial treatment group [27].

To curb tumor growth and disease progression, others have explored modulating both oncogene expression and anti-apoptotic signaling [23,24]. For example, Chowdhury et al. (2018) concentrated on increasing miR-122 activity to resensitize cells to chemotherapy and to silence oncogenic miR-21, which inhibits tumor suppressor genes [31]. To test this treatment’s ability to reverse chemotherapeutic resistance, miR-122 and antimiR-21 were loaded into biodegradable poly (lactic-co-glycolic acid) nanoparticles (PLGA-NP) and incubated with doxorubicin (DOX)-resistant and DOX-nonresistant HCC cells (Fig. 4). After treatment, DOX was found to be cytotoxic to cells at a lower dosage than prior to treatment, indicating resensitization of cells to DOX [24]. Moreover, post-treatment, HCC cells exhibited decreased anti-apoptotic factors and increased pro-apoptotic protein expression, and effects were greatest for combination treatments of miR-122 and antimiR-21 [24]. Additionally, expression of multiple drug resistance (MDR) protein was evaluated in DOX-resistant HCC cells since expression of MDR protein is upregulated in these cells [24]. Codelivery of miR-122 and antimiR-21 decreased MDR protein expression more than either miRNA alone, further supporting that miRNA combination treatment decreases DOX drug resistance. Construct efficacy was validated in vivo with MB-US mediated delivery into DOX-nonresistant and DOX-resistant subcutaneous xenograft mouse models. Both DOX-nonresistant and DOX-resistant HCC models revealed increased apoptotic behavior up to 6-fold in combination treatments as compared to DOX or miRNA delivery alone [24]. Similarly to in vitro results, treatment in xenograft models resulted in decreased expression of anti-apoptotic proteins in both DOX-nonresistant and DOX-resistant models [23,24]. In 2018, this same group noted that repetitive chemotherapeutic delivery cycles of the same treatment significantly improved results in a longitudinal study [24]. In this study, tumor shrinkage and survival rate, which were not reported in their first investigation, were significantly improved following combination treatment. These data are encouraging and support the success of consecutive, combination treatments for HCC therapy.

Figure 4: Mechanisms of MB-FUS extravasation of PLGA-NPs loaded with miR-122 and antimiR-21.

miR-122-and antimiR-21-loaded NPs enter HCC cells through endosomal pathways and undergo endosomal escape, allowing upregulation of miR-122 and downregulation of miR-21 [24].

Colorectal Cancer:

Following liver cancer, colorectal cancer (CRC) is the fourth most common cause of cancer-related deaths, affecting 4–5% of the population [32]. Current treatments typically include tumor resection followed by chemotherapy and/or radiation therapy [33]. However, surgery is not always a viable option due to tumor location, patient age, and potential underlying conditions, leaving approximately 40% of CRC patients without any therapeutic options [33]. Despite the widely understood network of signaling pathways involved in CRC, genetic mutations, and abnormal miRNA expression contributing to tumorigenesis [32], there is a stark lack of research observing anti-cancer gene delivery in the CRC space.

In this section, we will focus on one study observing MB-FUS gene delivery to subcutaneous colorectal cancer xenografts. Treating orthotopic colorectal tumors with FUS is challenging due to gas in the bowels, which blocks ultrasound transmission into tumor tissue [34]. Additionally, natural bowel movements may also complicate focusing treatment on the tumor itself [34]. To our knowledge, this is the only MB-FUS colorectal cancer study, highlighting the extent of these limitations. In their study, Wang et al. (2015) optimized in vivo FUS parameters for selective targeting and deep tumor penetration of fluorescent semiconducting polymer nanoparticles (SPN) to visualize tumor distribution. Selected parameters were then applied to deliver a similarly sized therapeutic: miR-122 loaded pegylated poly(lactic-co-glycolic acid) NPs (PLGA-PEG-NP), an FDA approved nanocarrier [7]. miR-122 NPs were successfully delivered to tumors with minimal tissue damage revealing biocompatibility of the regimen [7]. Nevertheless, the study lacked any therapeutic analysis of the impact of miR-122 delivery to the tumor such as tumor cell apoptosis, tumor volume, or subject survival. Although therapeutic impact was not investigated, a protocol to deliver therapeutics with an FDA-approved nanocarrier to colon tumor tissue was developed, thus creating a platform for downstream analysis of miR-122 delivery.

Pancreatic Cancer:

For pancreatic cancer patients, early-stage disease is often asymptomatic [35]. Tumors are most commonly diagnosed when resection is no longer feasible due to locally advanced cancer or metastasis, comprising about 80% of clinical cases [36]. With increased disease risk unattributable to one sole cause, this cancer is one of the most fatal human cancers, with about 9% living 5 years post-diagnosis [36]. Recourse apart from resection is comprised of radiotherapy and/or chemotherapy, most commonly with gemcitabine or fluorouracil [37]. Unfortunately, treating this deadly disease has not significantly progressed, as chemoresistance and poor drug permeability remain difficult challenges.

The treatment of pancreatic cancer has been approached from a combination drug-gene delivery perspective [38]. In addition to gemcitabine (GEM), miR-21 inhibitor (miR-21i)-loaded dendrimers were delivered on gold NPs to subcutaneous pancreas tumors. Associated with downregulation of tumor suppressors phosphatase and tensin homologue (PTEN) and downstream Akt, miR-21 is commonly upregulated in human pancreatic cancer, and miR-21i combined with GEM was found to impede pancreatic cancer cell growth in vitro. FUS-mediated delivery of miR-21i and GEM yielded higher permeability in a subcutaneous tumor model compared to the GEM and miR-21i monotherapies, indicating a benefit to this combination therapy. Furthermore, there was found to be increased tumor growth control 3 weeks after treatment, a significant survival advantage, and a higher percentage of apoptotic cells. Combination treatment far surpassed that of GEM alone, due to synergistic effects of the three therapeutic elements: FUS, GEM, and miR-21i. Although the authors did not examine the mechanistic underpinnings of increased therapeutic accumulation, it is clear that FUS played a role in improving GEM and miR-21’s combined therapeutic efficacy, providing insight as to how similar steps may be taken in the clinical space.

Hormone-related cancers:

Hormone-related cancers, such as breast cancer (BC), ovarian cancer (OC), and prostate cancer (PC) account for most cancers diagnosed in the United States, which comprise about 43% of all cancers diagnosed in women and 24% in men [39]. A variety of pro-proliferative hormone signals help spur hyperplasia in these cancers [40]. For this reason, inhibition of hormones and their receptors is often efficacious in treating hormone-dependent primary BC, OC, and PC tumors. Unfortunately, latent resistance from lack of hormone receptors or acquired resistance often hinder therapeutic efficacy, motivating clinical innovation to bypass these hormone pathways or to address them differently. Recently, FUS-mediated gene delivery approaches in preclinical murine models have targeted two main therapeutic avenues: hormone-dependent processes and targets associated with other cancer hallmarks. Addressing these pro-cancer pathways’ genetic underpinnings by up- or downregulating certain genes may achieve therapeutic effects alternative to small molecule inhibitors. Like many other solid tumors, this subgroup of tumors displays heterogeneity in cancer drivers, addressed herein with various gene therapy approaches.

Hormone dependency-associated approaches

In this subsection, we will discuss gene signatures associated with hormone dependency and FUS-mediated tumor growth control. Bai et al. (n.d.) investigated the impact of knocking down ABCG2, a drug efflux transporter that is both overexpressed in some breast cancers and associated with hormone dependence [41]. This work centered on the effects of DOX treatment on tumors from the MCF-7 human estrogen receptor (ER)+ cell line. The authors found that, in both DOX-sensitive and DOX-resistant tumors, delivery of ABCG2-siRNA with FUS and cationic MBs (CMBs) resulted in improved tumor growth control compared to delivery without FUS, indicating potential resensitization to DOX. Although a later publication [42] posited that drugs of the anthracycline class, including DOX, are poor substrates for this efflux transporter, this study supported the strategy of attenuating therapeutic resistance in hormone-related tumors. The knockdown of ANT-2, also associated with hormone dependence, was investigated by Park et al. (2015) in MDA-MB-231 triple negative breast cancer (TNBC) human xenografts [43]. ANT-2 regulates cell metabolism by converting mitochondrial ATP to cytosolic ADP, and its upregulation has been found to be associated with hormone dependency [44]. They employed poly-ethyleneimine (PEI), ANT-2 shRNA, and MBs, taking advantage of PEI’s pro-endocytic and cationic properties [45,46] to enhance shRNA uptake and protect shRNA from degradation. The subcutaneous TNBC tumors treated with MBs, PEI, ANT-2 shRNA and FUS were significantly smaller at day 55 and survival was prolonged as compared to scrambled shRNA controls. Together, these two publications advance strategies to combat resistance mechanisms indirectly related to hormone dependence.

Targeting hormones and their receptors

In this subsection, we will focus on treatments that target hormones and their receptors. Such approaches are appealing because many breast cancers exploit estrogen for proliferative signaling [40], and many ovarian cancers and early prostate cancers similarly depend on estrogen and androgen, respectively. Wang et al. (2014) have addressed the androgen receptor (AR) directly, employing siRNA, either conjugated or unconjugated to NBs, to knock down AR expression in androgen-sensitive C4–2 human xenografts [47]. They reported significant AR knockdown at the transcript and protein levels when animals were treated with AR siRNA conjugated to NBs and FUS, as well as significant tumor growth control at 3 weeks. Similarly, Zhao et al. (2018) targeted FOXA1 in an ER-dependent BC model, which regulates chromatin condensation around ER’s nuclear receptor binding site, affecting ER expression [48]. Using cationic porphyrin MBs to protect FOXA1 siRNA and facilitate reactive oxygen species (ROS) production, gene delivery and photodynamic therapy (PDT) were leveraged to knock down ER’s proliferative function and induce cell stress. Upregulation of pro-apoptotic caspase-3 and reduced cell proliferation were observed in vitro and hypothesized to facilitate similar FOX1A protein downregulation in vivo, correlating with ROS production, a concept illustrated in Figure 5. Additionally, tumor growth was suppressed at 3 weeks for mice treated with FUS, FOXA1 siRNA, and PDT compared to each monotherapy. Together, these two studies offer evidence that FUS-enhanced RNAi and delivery of pro-apoptotic agents may be enough to prevent primary tumor growth in hormone-sensitive tumor models. Given the high proportion of BC, OC, and PC patients to which this may apply, these approaches may have improved clinical impacts compared to current therapeutic options, providing insight to circumvent hormone dependence.

Figure 5: A depiction of MB-FUS delivery of FOX1A siRNA in combination with PDT to BC xenografts.

A) Cationic porphyrin-grafted NP-MB complexes coated with siRNA comprised the delivery vehicle. B) Contrast-enhanced ultrasound (CEUS) technology forms the base of low-frequency ultrasound (LFUS), whose application allows FOX1A knockdown via improved siRNA delivery and augmented cell stress and death after laser stimulation of porphyrin-grafted MBs [48].

Targeting EGFR family proteins

In addition to hormones and their receptors, a large subset of BC patients (~15–20%) express HER2 [49,50] and other epithelial growth factor receptor (EGFR) family proteins, which may also serve as viable therapeutic targets. EGFR protein phosphorylation induces several pro-proliferation pathways, including PI3K/Akt and Ras/Raf/MEK/ERK, contributing to hyperplasia [51]. Thus, anti-HER2 therapies have been devised and are currently used either to block HER2 function or to target other drugs to HER2+ cells [52]. However, this can result in acquired resistance to HER2 targeting, similar to hormone-related resistance. Song et al. (2019) have argued that anti-HER2 resistance could be addressed through RNAi of epithelium-specific Ets transcription factor 1 (ESE-1), a positive HER2 effector [53]. ESE-1 siRNA delivered in combination with CMB-FUS was successful in reducing tumor volume over the course of two weeks compared to controls. Although TNBC lacks HER2 expression, other EGFR family proteins exist that elicit similar proliferative signaling cascades [54], potentially serving as therapeutic targets. Jing et al. (2016) have treated human TNBC cells with NB-FUS and EGFR siRNA to directly inhibit this overexpressed cell surface protein [55]. After confirming EGFR knockdown and decreased cell proliferation in vitro, the construct was delivered to TNBC subcutaneous mouse tumors, which subsequently exhibited significant tumor inhibition over 20 days. Despite limitations of sustained RNAi response, it is encouraging that some therapeutic efficacy was achieved, potentially bolstering the case for repeated delivery of RNAi agents.

Hormone-independent approaches

The publications in the previous section addressed therapeutic targets isolated to BC, OC, and PC, especially related to hormone receptors and the EGFR protein family. Here, we will address gene therapy endeavors potentially applicable to a range of solid tumors, including cytokine-based immune modulation, apoptosis induction, hypoxia regulation, and proliferation control. Although the focus of these strategies is wider than the hormone-related targets above, it is important to note that circumventing hormones and their receptors may provide headway for patients with TNBC or cancers resistant to hormone-based strategies. In the following paragraphs, we discuss gene delivery in BC, OC, and PC related to (1) immune modulation, (2) induction of apoptosis, (3) regulating hypoxia, and (4) curbing proliferation.

Gene immunotherapies

Immune cells are influential players in the TME, dictating how and if tumors are recognized by the body and respond to immunomodulatory treatments [56]. Intratumoral immune cells consist of a variety of cell types, including cytotoxic T cells, helper T cells, macrophages, dendritic cells (DCs), natural killer (NK) cells, myeloid derived suppressor cells (MDSCs), and T regulatory cells (Tregs) [56]. In this section, we review studies which leverage host immune response to augment FUS-enhanced tumor control, two of which are focused on IL-12 and IL-27 plasmid delivery. Activated DCs produce IL-12 and IL-27, which increases IFN-γ production and thus increases naive CD4+ T cell proliferation, NK cell stimulation, and CD8+ T cell maturation. Both Suzuki et al. (2010) and Zolochevska et al. (2011) attempted to upregulate IL-12 in OC and IL-27 in PC, respectively, using MB-FUS to improve delivery efficiency. These approaches aim to either increase the likelihood of intratumoral immune infiltration or antigen presentation, both of which may generate an anti-tumor response. Suzuki et al. (2010) performed selective depletion of NK or T cells to determine if therapeutic impact depended on individual immune cell subtypes [57], a common technique used to investigate roles of immune cells in specific pathologies. It was found that delivery of a plasmid IL-12 gene construct with MB-FUS increased infiltrating T cell counts in addition to intratumoral perforin, a cytolytic protein secreted by NK and CD8+ T cells. They noted that NK cell depletion had little effect on tumor growth control but that T cells were essential, highlighting their importance in immune cell-mediated tumor response. Further, they reported that 5 consecutive doses of the therapy suppressed tumor growth in a highly metastatic murine OV-HM model for a longer period, illustrating the effect of repeated gene administration on more sustained tumor growth control. Zolochevska et al. (2011) discovered that IL-27 plasmid delivery with MB-FUS was sufficient in reducing subcutaneous transgenic adenocarcinoma of the mouse prostate 2 (TRAMP2) tumor volume by threefold when treating at day 0, 2, 4, and 6 [58]. Corroborating Suzuki et al.’s (2010) findings in highlighting the importance of T cell function in tumor growth control, qPCR analysis indicated upregulation of genes involved in T cell activation (Sit1, Irf4, Cd8a, Cd8b1), proliferation (IL-10), and differentiation (IL-27, Irf4, Jag2).

Interferons (IFNs) are a subset of cytokines which are secreted in response to viral invaders to promote their eradication [59]. Because this robust immune response can be mounted naturally by the body, researchers have postulated that its induction by cancer cells may spur immune attack on tumors [60]. Ilovitsh et al. (2020) endeavored to upregulate IFN-β production with IFN-β plasmid (pIFN-β) delivery, targeting one of two syngeneic subcutaneous BC tumors in a bilateral tumor model [61]. This was motivated by the fact that IFN-β is often used to modulate T cell responses and has been shown to suppress tumor growth via CD8+ T cell-related mechanisms [62]. pIFN-β delivery with targeted microbubbles (TMBs) demonstrated enhanced macrophage and CD8+ cell presence, potentially implicating a cytotoxic immune response due to increased IFN-β expression, a process highlighted in Figure 6. In addition to these findings, the group found greater tumor growth control at day 7 after a singular treatment, both in treated and distal tumors compared to controls. Despite less tumor growth control elicited in the distant tumor, pIFN-β delivery to the treated tumor still curbed contralateral tumor growth to some extent. Notably, applying anti-PD-1, an immune checkpoint inhibitor, increased distal tumor control when delivered with pIFN-β and FUS, likely a result of greater T cell activation, showing the potential of combination therapies.

Figure 6: Anti-PD-1 and MB-FUS delivery of IFN-β plasmid combination therapy.

MB-FUS-mediated delivery of IFN-β plasmid in concert with intratumoral anti-PD-1 injection increased the number of tumor-associated macrophages and cytotoxic CD8+ T cells and decreased the number of tumor cells [61].

These three studies highlight the importance of effector T cell infiltration and function in combating solid tumor growth with upregulation of IL-12, IL-27, and IFN-β. Effector T cells were involved with tumor regression in all cases, suggesting that anti-cancer therapeutic success depends in part on tumor immune infiltrate in hormone-related cancers. Additionally, targeted gene delivery with FUS may serve as a safer alternative to systemically delivered cytokine therapy, as previous systemic delivery approaches have generated immune responses with adverse side effects in patients [63]. By localizing cytokine production through selective transfection, systemic toxicity may be avoided and cytokine gradients important to immune cell homing may be enhanced.

Pro-apoptotic gene therapies

Yet another subset of studies has leveraged FUS to directly elicit apoptosis. Systemically administered apoptotic agents may result in off-target toxicity, hence localized gene therapy-mediated apoptosis is an attractive strategy. In solid tumors, cells often develop mutations that result in aberrant function and a reduction in apoptosis. Common culprits include p53, referred to as the “guardian of the genome”, whose expression can halt cell cycling and induce apoptosis in damaged cells, and caspases, which affect cell death at the end of apoptotic signaling cascades [64].

In 2017, Devulapally et al. demonstrated tumor growth control after suicide gene delivery in a TNBC subcutaneous murine model [65]. They employed a thymidine kinase-nitroreductase (TK-NTR) construct with a prodrug, aiming to selectively elicit cell death in tumor tissue driven by survivin, a tumor-specific promoter, converting the prodrug to a cytotoxic compound. Complexing TK-NTR gene plasmids with PEI/PLGA NPs increased gene uptake, and delivery of this construct with MB-FUS enhanced intracellular plasmid accumulation in vitro. In vivo treatment of the TNBC subcutaneous tumor demonstrated that this regimen delivered twice over two weeks resulted in significant tumor control, more so than the gene construct with FUS or the prodrug alone. Although there were no ex vivo apoptosis analyses, this group was able to show brief tumor growth control in a tumor refractory to hormone or EGFR-based therapies.

Taking a different approach, Greco et al. (2010) demonstrated therapeutic efficacy of a replication-incompetent IL-24-encoding adenovirus [66]. IL-24 is postulated to upregulate beclin-1 and p53 upregulated modulator of apoptosis (PUMA), mediators of autophagy and apoptosis, respectively [66]. They posited that beclin-1 and PUMA upregulation could overcome beclin-2 or beclin-XL-mediated resistance, as beclin-2 and beclin-XL bind to beclin-1 to inhibit autophagy in cancer cells. Their gene construct design centered around adenovirus replication driven by the progression-elevated gene-3 (PEG-3) promoter, which is associated with malignant transformation of cells. Employing an IL-24 viral vector, they aimed to transduce one tumor in a bilateral hormone-insensitive PC xenograft model with FUS. This intervention resulted in increased tumor growth control in treated tumors, in untreated contralateral tumors, and in autophagy-resistant, beclin-XL-overexpressing tumors, with no tumor regrowth or metastasis at 3 months post-treatment. This indicates that, at least in nude mice, this FUS-mediated IL-24 gene therapy can circumvent anti-autophagy signaling, resulting in sustained tumor growth control in this model.

Lastly, Horie et al. (2011) concentrated on the TNF-α cytokine and the effects of its upregulation in TNBC tumors [67]. TNF-α is secreted by macrophages and binds to receptors on a variety of immune and endothelial cells; however, it is currently unknown if it elicits more pro- or anti-tumor effects [68]. Treatment using this cytokine has been stymied by significant toxicity due to off-target effects of inflammation and caspase activation, resulting in apoptotic upregulation in noncancerous cells. Immunocompromised SCID mice were used in this group’s experiments and were treated on days 2, 4, 7 and 9 after inoculation with NB-US and TNF-α plasmid. Repeated NB-US plasmid delivery resulted in significantly smaller tumors compared to control and systemically TNF-α-treated mice. Animals treated with NB-US and plasmids had decreased normalized vessel area, and decreased CD31 expression, an endothelial cell marker, indicating lower blood vessel density in tumors after treatment. Together with an upregulation of p53 and caspase-3 mRNA, it may be deduced that the treated tumors were susceptible to TNF-α. However, it would be interesting to see if this is repeatable in p53-mutant tumor lines and in immunocompetent models given the high incidence of p53 mutations in solid tumors and the importance of adaptive immune cells in antitumor responses. From this section, we can deduce that apoptosis-inducing gene delivery should be validated in immune-competent animals and done so by metrics other than tumor size. Regardless, the three studies discussed herein demonstrate the potential for a wide range of pro-apoptotic, anti-cancer approaches.

Anti-angiogenic strategies

Hypoxia and its effects on both BC and OC tumor angiogenesis have also been studied in the context of US-mediated delivery of genetic material. Since blood vessels bring nutrients and provide gas exchange to and from tissues, oxygen and vasculature are directly linked to tumor growth [69]. The harsh environments that cancers cells can withstand allow tumors to survive despite fewer nutrients in the TME and poor tumor vasculature, thus limiting therapeutic delivery [69]. Groups investigating FUS-mediated gene therapy applications explored different approaches. One involved knocking down hypoxia-inducible factor 1-alpha (HIF-1α), whose upregulation in hypoxic environments helps cells function despite low oxygen levels. Another employed vascular endothelial growth factor (VEGF) downregulation, a cytokine which promotes blood vessel formation. It is still unclear whether it is more beneficial to reduce or augment intratumoral hypoxia, as an ideal equilibrium between cell starvation and angiogenesis within the tumor is challenging [69]. Nonetheless, multiple approaches to modulate blood vessels in tumors have been pursued in search of potential therapeutic effects.

HIF-1α stimulates angiogenesis and is highly degraded under normoxic conditions. However, it is not degraded under hypoxic conditions, resulting in pro-angiogenic signaling [30]. In concert with VEGF upregulation, HIF-1α is postulated to contribute to the irregular, leaky vasculature present in many solid tumors. Sun et al. (2018) used a PDT approach to deliver HIF-1α siRNA with porphyrin MBs, attempting to downregulate hypoxic signaling by knocking down HIF-1α and leveraging PDT to enhance ROS-induced tissue damage and cell death [70]. Additionally, treatment with HIF-1α siRNA, MBs, US, and PDT resulted in significant tumor growth control compared to individual treatment components, highlighting the power of combination treatment. HIF-1α downregulation resulted in decreased cancer cell proliferation marked by Ki67 downregulation and pro-apoptotic caspase-3 upregulation, indicating cell death. Florinas et al. (2013) employed a different approach, opting for VEGF siRNA delivery to OC cells with the help of cationic polymer-coated MBs [5]. In an in vitro experiment, the group demonstrated that serum proteins absorb nucleic acids, evidenced by 0% uptake of siRNA in serum-enriched media despite the use of MBs and FUS. Cationic MBs reduce systemically administered genetic material’s susceptibility to protein adsorption and prolong nucleic acid circulation time [5]. siRNA delivered with CMBs resulted in smaller tumor volume compared with controls over a 10-day period after delivery. It would be interesting to note how downregulating VEGF might impact selection of hypoxia-resistant cells in a longer-term experiment. Indeed, two different gene therapy strategies proved to be useful in limiting pro-angiogenic signaling in hormone-related tumors, demonstrating the anti-cancer potential of FUS-mediated gene therapy.

Anti-proliferation approaches

A final subset of hormone-related cancer therapies focuses on delivering proliferation inhibitors to control tumor size and extend survival, with many of these inhibitors exhibiting secondary apoptotic or drug sensitizing effects. Liu et al. (2021) used a dual-targeted AKT2 shRNA to decrease signaling of the PI3K/AKT pathway and slow cancer cell proliferation, aiming to further improve delivery with MB-FUS to MCF-7 xenografts [71]. Both iRGD peptide and C-C chemokine receptor type 2 (CCR2) were involved in the targeting strategy, the former incorporated into the MB shell for selective endocytosis aimed at αVβ3 integrin binding and the latter a commonly upregulated receptor on breast cancer cells. Demonstrating that there was increased cancer cell shRNA uptake, a larger number of cells in G1, and a smaller number of cells in S phase, the group concluded that cell cycle interruption had occurred in vitro. Following these experiments, increased shRNA accumulation by FUS was confirmed in vivo, with iRGD and CCR2 combination targeting resulting in better tumor growth control 15 days after treatment. These results combined with high cell apoptosis and cell cycle arrest in tumors treated with the iRGD- and CCR2-targeted MBs highlight the utility of precise targeting strategies.

Wu et al. (2018) developed a unique PC cell targeting strategy of A10–3.2 aptamer-conjugated NBs to improve anti-proliferative siRNA transfection [72]. Aptamers are short single-stranded oligonucleotides which bind to cell surface biomolecules, making them a viable targeting ligand [73]. The A10–3.2 aptamer binds to prostate-specific membrane antigen (PSMA), a cell surface protein upregulated in hormone-dependent and metastatic PC. Aptamer-conjugated NBs were loaded with FoxM1 siRNA, whose overexpression in PC promotes abnormal cell proliferation and whose knockdown can slow proliferation and cause PC cell apoptosis. In a subcutaneous xenograft, the group showed tumor size control, likely boosted by FoxM1 knockdown, downregulation of pro-migratory and metastatic E-cadherin, and an increase in apoptotic cells. More recently, Liu et al. (2020) targeted SIK2, a mitotic spindle development protein implicated in OC metastasis in conjunction with inhibiting miR-21 [74]. They reported that greater than ⅓ of an ovarian cancer clinical cohort had SIK2 expression in their tumors, clinically validating SIK2 as a therapeutic target. Delivery of these nucleic acids in a subcutaneous xenograft model was facilitated by PLGA NPs injected concomitantly with MBs, yielding significant relative tumor shrinkage 3 weeks after treatment as well as a survival advantage at 60 days compared to treatment controls. This therapeutic effect was supported by an increase in apoptotic cells and an increase in cell cycle arrest. These anti-proliferative approaches underscore the wide possibility of therapies that may function to undermine solid tumor progression, potentially improving clinical outcomes in patients resistant to current chemo- and hormone-targeted therapies.

Skin Cancer:

Skin cancer is the most common form of cancer worldwide. There are 3 main types of skin cancers: squamous cell carcinoma, basal cell carcinoma, and melanoma [75]. While chemotherapeutics exist, their efficacy is limited by cancer’s characteristic susceptibility to drug resistance and treatment side effects [74]. Furthermore, anti-PD-1 immunotherapies, which are typically used to treat melanoma, sometimes present with adverse autoimmune side effects [76]. With increasing knowledge of signaling pathways involved in skin cancer, gene therapy has emerged as a promising therapeutic for highly selective and specific anti-cancer therapy. This next section outlines recent progress in FUS-mediated skin cancer gene therapy.

Squamous Cell Carcinoma

Similar to other cancers, oncogenic and anti-apoptotic factors are upregulated in squamous cell carcinoma (SCC) and promote tumor growth [77,78]. SCC has been treated with small molecule inhibitors specifically targeting EGFR, which is commonly implicated in tumor progression [78]. However, these inhibitors are unable to completely block EGFR signal transduction [78]. With the advent of gene therapy, EGFR inhibitors have often been supplemented with other treatment modalities, like RNAi, to further modulate aberrant oncogene expression in mouse models [78]. Originally, gene therapies were administered with direct intratumoral injection, which hindered this approach’s translation to the clinic due to its invasive nature [79]. Carson et al. (2012) introduced a novel non-invasive, non-viral EGFR-targeted siRNA treatment modality for SCC tumors [78]. This study was also the first to investigate consecutive MB-FUS siRNA delivery treatments. With this protocol, tumor growth was significantly inhibited and tumor cell apoptosis was enhanced compared to controls [78]. Furthermore, the magnitude of tumor inhibition in this study was greater than that of intratumorally injected therapeutics, supporting MB-FUS siRNA as a safer and more effective treatment modality [78].

Rather than targeting a single oncogenic receptor, Kopechek et al. (2015) investigated silencing of an upstream signal transducer associated with drug resistance of many oncogenic factors: signal transducer and activator of transcription 3 (STAT3) [79]. STAT3 inhibition affects multiple downstream oncogenic sites and was hypothesized to be more efficient than previously studied EGFR-siRNA, which targets a single receptor [79]. In order to silence STAT3, a cyclic STAT3 decoy structure was developed that binds to STAT3 protein, preventing signal transduction. They used MB-FUS to facilitate non-invasive and non-viral delivery of STAT3 decoy on cationic lipid-coated MBs in a SCC murine tumor model. An added benefit of this construct is its therapeutic stability, as DNA (i.e. STAT3 decoy) is inherently more stable than RNA (i.e. EGFR-siRNA) and complexing the decoy to MBs further protects it from degradation [80,81]. Furthermore, the mechanism of silencing a preexisting protein by blocking its receptors rather than implementing RNAi, which may cause more transient gene silencing, also offers a potentially more efficacious therapeutic technique [77,79]. Studying the delivery of this construct in vivo, the authors identified reduced expression of target genes downstream of STAT3, including beclin-XL and cyclin D1 (Cyc-D1) [79]. Tumor volume was also notably reduced in comparison with previous studies and the use of MBs significantly increased the delivery of the therapeutic [79]. To improve physiological and clinical relevance, these experiments were later repeated using human SCC xenografts. Similar to their previous study, they found that STAT3 downstream target gene expression of beclin-XL and Cyc-D1 was downregulated and tumor growth was significantly slowed [77,79].

Han et al. (2020) developed a nanocomplex aimed at overcoming the challenges of simultaneously encapsulating nucleic acids and chemotherapeutic agents [82]. RNA hydrogel nanoparticles were complexed to PTX-loaded MBs, which can disrupt surrounding cell membranes and extracellular matrix (ECM) upon FUS-induced MB oscillation. Figure 7 illustrates the formulation, application, and proposed FUS-mediated effects of the construct in vivo. In many studies, therapeutic delivery and efficacy are hindered by poor agent accumulation due to EPR effect limitations and shallow tumor penetration due to dense ECM [82]. This construct successfully overcame both of these obstacles with MB-FUS-induced cell sonoporation and loosening of the surrounding ECM [82], enabling deep and efficient tumor penetration of PTX and siRNA in vivo.

Figure 7: Formulation and mechanisms of siRNA-NP-PTX-MB vascular and ECM disruption for enhanced siRNA and PTX tumor penetration.

A) Fabrication of RNA hydrogel nanoparticles. B) Schematic of siRNA-loaded NPs and PTX-loaded MBs. C) Schematic demonstrating the delivery of the siRNA-NP-PTX-MB complex in vivo. Upon FUS exposure at the tumor site, the complex oscillates and loosens the ECM, facilitates sonoporation of the vasculature, and enhances delivery of siRNA and PTX into the tumor tissue [82].

Melanoma

Currently, reports of in vivo FUS-mediated gene therapy treatments for melanoma are lacking in the literature. In this section, we will focus on two studies: one study regulating macrophage activity in primary tumors and one validating agent delivery in secondary tumors. C-Type Lectin Domain Family 4, Member E (Clec4e, also called Mincle) is expressed by inflammatory macrophages and has been implicated in progression of various cancers because of its involvement in inflammatory cytokine production, thus making it a desirable therapeutic target [83]. An absence of Mincle-targeting therapies motivated Li et al.’s (2020) pursuit and eventual discovery of a novel Mincle/Syk/NF-kB signaling circuit regulating pro-tumoral activities of tumor activated macrophages (TAMs). Mincle plays a role in suppressing M1 phenotypes and enhancing M2 phenotypes of TAM, contributing to tumor growth and immunosuppression. To silence Mincle, Li et al. (2020) used MB-FUS to deliver Mincle-targeted shRNA to syngeneic melanoma mouse models. This treatment significantly reduced Mincle protein expression and tumor volume, correlating successful gene silencing with tumor growth inhibition [83]. Furthermore, IL-6, which supports tumor progression, was significantly downregulated in treated mice [83]. Additionally, Curley et al. (2020) validated MB-FUS delivery of BPNPs to secondary, metastatic murine melanoma in the brain and noted that FUS can be manipulated to allow penetration of BPNPs into the entire tumor and surrounding tissue [13]. Though therapeutic effects of anti-cancer treatments were not reported, this study laid the groundwork for future experiments delivering and monitoring effects of therapeutics, like nucleic acids.

Lung Cancer:

As the second-most common cancer in men and third-most common in women [84], lung cancer is annually diagnosed in about 1.5 million patients worldwide. When diagnosed in early-stage disease, 5-year survival is around 50–80% [85], with 30–55% of patients having undergone resection experiencing recurrence [86]. This reflects that current surgical resection, chemotherapy, and radiotherapy clinical standards are not sufficient to cure patients, which opens the door for FUS-based gene therapies to potentially improve upon.

To our knowledge, there is only one study centered around FUS-related gene delivery in vivo in lung cancer, which identified a significant difference between Minlce expression in TAMs and cancer cells in human non-small cell lung carcinoma (NSCLC) tumors [83]. Delivery of shRNA with FUS and MBs was performed over three treatment sessions in a Lewis cell carcinoma syngeneic murine model and analyzed at day 22 post-inoculation. This therapy resulted in significantly smaller tumors, lower levels of Mincle mRNA, and less serum IL-6 compared to controls. Since IL-6 is associated with increased TAM activity, the concurrent reduction of intratumoral Mincle expression may indicate a particular mechanism implicating TAMs, Mincle, and IL-6 expression in both NSCLC and melanoma murine models [83]. This research also evokes the importance of regarding immune cell infiltration and modulating their behavior.

Future Directions:

Throughout this review, we have outlined recent progress of FUS-mediated gene delivery in solid tumors and highlighted findings potentially foundational to clinical therapies. Despite this, further improvements are imperative prior to further translation of these therapies. Successful therapeutic delivery is hindered by intra- and intertumoral heterogeneity preventing adequate treatment, which may be overcome in some applications by combination treatments [19]. Additionally, some molecular targets are not unique to tumors, increasing the possibility of negative off-target effects [19]. Thus, further improvements may be realized through identification of unique tumor cell targets, which can inform surface modifications of delivery vectors for improved selective targeting of the tumor. Additionally, tissue- or cell-specific promoter regions employed during plasmid design may restrict which transfected cells respond to the delivered nucleic acids, as otherwise off-target effects may occur. Rather than targeting cancer cells with specific cell surface receptors, recent work has emerged pioneering the use of MB-FUS to selectively target and transfect endothelial cells in the cerebral vasculature, termed sonoselection [87]. We suggest that a similar effect might be elicited with MB-FUS applied to tumors, selectively transfecting tumor vasculature. This may have exciting biological implications, potentially permitting delivery of gene constructs explored in this review specifically to tumor blood vessels.

Looking ahead, the further incorporation of transcriptomic and proteomic profiling may be useful for identifying potential molecular targets, predicting patient responses, and giving insight on mechanisms underlying improved delivery [88]. Transcriptomic analyses of FUS-mediated plasmid delivery to a naïve mouse brain have been undertaken by Mathew et al. (2021), indicating preferential uptake by certain cell types [89]; this may also be the case in solid tumors. Gene expression has been shown to change over time in response to FUS [14,89,90], therefore it may be possible to predict if or when therapeutics should be readministered, and if FUS changes these signatures in a way that may alter therapeutic efficacy.

Many of the gene therapy mechanisms detailed throughout the review result in transient gene up- or down-regulation, which hinders sustained therapeutic efficacy. CRISPR gene editing provides an improved avenue for more robust MB-FUS gene delivery, capable of permanently adding or deleting genes implicated in cancer progression. While CRISPR plasmids are subject to degradation like other nucleic acids, they produce longer lasting effects as they modify the cellular genome rather than transiently silencing or amplifying gene expression [4]. Ribonucleoproteins (RNPs), a form of CRISPR technology, are made up of Cas9 proteins complexed to guide RNA. These complexes forgo the need for translation and transcription of the two components, respectively, simplifying CRISPR gene editing. RNPs are also smaller than CRISPR plasmids, potentially allowing for more efficient FUS-mediated gene delivery [4]. For these reasons, it would be interesting to observe how therapeutic efficacy of CRISPR/RNP gene delivery differs from other gene delivery strategies covered in this review.

Despite therapeutic effects seen in most studies discussed in this review, the problem of gene delivery efficiency remains, necessitating improvements in delivery. Cationic MB surfaces and PEI-coated MBs were two strategies employed across several studies, which aim to protect nucleic acids in systemic circulation and increase particle uptake through endocytosis, respectively [45,46]. Beyond these methods exist a host of more complex delivery systems, including gene carrying-NPs [91,92] and complexes of gene-loaded nanoparticles and MBs [82,92,93]. An example of a successful preclinical study with magnetic MB-FUS-assisted DOX delivery is shown in Figure 8, being one of many novel concepts that may be extrapolated to MB-FUS-mediated gene delivery [94]. Due to the heterogeneity of solid tumors, it is likely that no singular delivery vehicle formulation can be translated with the same efficacy across tumor types, highlighting an additional challenge of gene delivery strategies. Although further studies of these NPs are required before clinical translation, there are many preclinical studies demonstrating how NPs can improve gene delivery to tumors outlined in this review.

Figure 8: Magnetic liposome (ML)-conjugated MBs loaded with DOX improve tumor accumulation.

Magnetic guidance of ML-MBs in conjunction with FUS resulted in increased agent accumulation and tumor growth control [94].

Despite preclinical progress in gene delivery, scaling treatment from small animal models to humans also poses several challenges. One potential obstacle for some indications is US transducer focal size, which is often on the order of a grain of rice [95]. Since human tumors can be orders of magnitude larger than murine tumors, translation of these studies to humans would require more elaborate treatment plans. Additionally. the human circulatory system is much larger in volume than that of the mouse, requiring materials to be scaled up to human size. Although nucleic acid fabrication remains a costly process, recent demand by biotechnology companies during the COVID-19 pandemic has highlighted the necessity to further streamline this process, and many academic and corporate groups are working to address this problem of scalability [96].

In this review, we have outlined recent progress in FUS-aided gene delivery in solid tumors. Although several gene-based strategies show preclinical promise, future improvements remain to further augment therapeutic effects of these approaches. With a multitude of promising gene targets and an ever-improving understanding of FUS bioeffects, FUS-mediated gene delivery continues to show promise in combating solid tumors, and we are hopeful that current clinical trial activity in the FUS-mediated drug delivery space will pave the way for FUS-mediated gene delivery clinical trials for solid tumor therapy in the near future.

Highlights.

• Ultrasound can breach physical barriers to nucleic acid delivery to tumors

• Numerous gene carriers have been successfully combined with ultrasound pre-clinically

• Ultrasound-targeted nucleic acid delivery is applicable to multiple neoplasms

• Promising advances in treating metastatic disease with ultrasound have been made

• Progress to date in the ultrasound gene delivery field is comprehensively reviewed

Glossary

AFP

α-fetoprotein

AIF

apoptosis-inducing factor

AR

androgen receptor

BBB

blood-brain barrier

BC

breast cancer

BCL-2

B-cell-lymphoma-2

BTB

blood-tumor barrier

BPNP

brain penetrating nanoparticle

CCR2

chemokine receptor type 2

CEUS

contrast-enhanced ultrasound

Clec4e

C-Type Lectin Domain Family 4, Member E (Mincle)

CMB

cationic microbubble

CNS

central nervous system

CRC

colorectal cancer

CRISPR

Clustered Regularly-Interspaced Short Palindromic Repeats

Cyc-D1

cyclin D1

DC

dendritic cell

DOX

doxorubicin

ECM

extracellular matrix

EGFR

epithelial growth factor receptor

EPR

enhanced permeability retention

ER

estrogen receptor

ESE-1

Ets transcription factor 1

FUS

focused ultrasound

GBM

glioblastoma

GEM

gemcitabine

GM

glioma

GPC3

glypican-3

gRNA

guide RNA

HCC

hepatocellular carcinoma

HIF-1α

hypoxia-inducible factor 1-alpha

IFN

interferon

LFUS

low-frequency ultrasound

LPH

lipid-polymer hybrid

MB

microbubble

MB-FUS

microbubble enhanced focused ultrasound

MB-US

microbubble enhanced ultrasound

MDR

multidrug resistance

MDSC

myeloid derived suppressor cells

miRNA

microRNA

ML

magnetic liposome

MRI

magnetic resonance imaging

NB

nanobubble

NB-FUS

nanobubble enhanced focused ultrasound

NB-US

nanobubble enhanced ultrasound

ND

nanodroplet

NET-1

neuroepithelial transforming gene 1

NK

natural killer

NP

nanoparticle

NSCLC

non-small cell lung carcinoma

OC

ovarian cancer

PBPK

pharmacokinetic

PC

prostate cancer

PDT

photodynamic therapy

PEG-3

progression-elevated gene-3

PEI

poly-ethyleneimine

pHSV-TK/GCV

non-viral herpes simplex virus thymidine kinase/ganciclovir plasmid

PI3K

phosphoinositide 3-kinase

pIFN-β

interferon-β plasmid

PLGA-NP

poly (lactic-co-glycolic acid) nanoparticle

PLGA-PEG-NP

pegylated poly(lactic-co-glycolic acid) nanoparticles

PSMA

prostate-specific membrane antigen

PTEN

phosphatase and tensin homologue

PTX

paclitaxel

PUMA

p53 upregulated modulator of apoptosis

RNAi

RNA interference

RNP

ribonucleoprotein

ROS

reactive oxygen species

SCC

squamous cell carcinoma

SHH

sonic hedgehog

shRNA

short hairpin RNA

siRNA

small interfering RNA

SMO

smoothened

SPN

semiconducting polymer nanoparticles

STAT3

signal transducer and activator of transcription 3

TAM

tumor activated macrophages

TIMP3

tissue inhibitor of metalloproteinase 3

TK-NTR

thymidine kinase-nitroreductase

TMB

targeted microbubble

TME

tumor microenvironment

TNBC

triple negative breast cancer

TRAMP2

transgenic adenocarcinoma of the mouse prostate 2

Tregs

T regulatory cells

VEGF

vascular endothelial growth factor

VEGFR2

vascular endothelial growth factor receptor 2