Recent advances in non-invasive in vivo tracking of cell-based cancer immunotherapies

Abstract

Immunotherapy has been at the forefront of cancer treatment research in recent years due to an increased understanding of the immune system’s role in cancer and the substantial benefits it has demonstrated compared to conventional treatment methods. In particular, immune cell-based approaches utilizing T cells, natural killer (NK) cells, macrophages, and more have shown great potential as cancer treatments. While these treatments hold promise, there are still numerous issues that limit their clinical translation, including a lack of understanding of their mechanisms and inconsistent responses to treatment. Traditionally, tissue or blood samples are collected as a means of monitoring treatment progression. However, these in vitro diagnostics are invasive and provide limited information about the real-time status of the treatment or its long-term effectiveness. To address these limitations, novel non-invasive imaging modalities have been developed. These include optical imaging, X-ray computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET) and single-photon emission computed tomography (SPECT), and photoacoustic (PA) imaging. This review focuses on methods for tracking cell-based cancer immunotherapies using these in vivo imaging modalities, thereby enhancing real-time monitoring of their therapeutic effect and predictions of their long-term efficacy.

1. Introduction

The immune system’s role in cancer treatment has been extensively studied in recent decades, advancing our understanding of cancer biology and influencing new therapeutic strategies. Immunotherapy, which leverages a person’s immune system to fight cancer, represents a significant deviation from conventional treatments such as chemotherapy and radiation.¹ Immunotherapy has shown improvements in patient survival and quality of life as compared to these conventional treatments.² There are two types of cancer immunotherapy: immune checkpoint inhibitor therapy and immune cell-based therapy. Immune checkpoint inhibitors (ICIs) are monoclonal antibodies that block checkpoint proteins, preventing immune cell inactivation by cancer.³ This approach differs from cell-based therapy, which involves modification and activation of immune cells to improve their ability to target cancers.

Cell based therapies use immune cells to target malignancies. The immune cells can either be a patient’s own cells, an autologous cell transplant, or a donor’s cells, an allogeneic cell transplant. Immune cells have specific properties that make them uniquely suited for use in the treatment of cancer.⁴ T cells have the natural ability to locate and target tumor cells as a part of their immune function, which makes them ideal for use in cell-based therapies. This innate ability to home to the tumor site can be exploited in the use of tumor-infiltrating lymphocyte (TIL) therapy, in which a patient’s own T cells are expanded outside the body and reinfused into the patient.⁵ NK cells have their own set of tumor-targeting properties, distinct from that of T cells.⁶ These cells are recruited to the tumor site by chemokines that are created by immune cells in the tumor microenvironment (TME).⁷ Unlike T cells, NK cells are able to recognize targets without the use of the major histocompatibility complex I (MHC I). This is useful in the case of cancers that downregulate expression of the MHC I complex as a method of immune evasion.⁷ Myeloid cells, such as macrophages and monocytes, can also be used in cell therapies. Macrophages and monocytes are unique in that they can act in a pro-tumor or anti-tumor manner depending on their phenotype. Since they are an innate part of the TME, myeloid cells have high retention at the tumor site, where they can be induced to an anti-tumor state.^8,9 Dendritic cells are antigen presenting cells, and they play a key role in the activation of the immune system. This makes them an ideal target for immunotherapies, as they target tumors themselves in addition to activating other cells that can target tumors.¹⁰

The ability of immune cells to target cancer cells can be enhanced using various bioengineering approaches, including direct modification of the cell or by inducing cells to target tumors using external stimuli. Immune cells like T cells and NK cells have been engineered with modified receptors such as T-cell receptors (TCR) and Chimeric Antigen receptors (CAR), which enables better recognition of cancer antigens, resulting in immune cell activation against tumor cells.¹¹ Using TCR enables T cells to recognize both surface and internal proteins through antigen presentation by an MHC complex. However, tumor cells are able to downregulate the expression of MHC complexes on their surface, which enables them to evade immune detection.¹² CARs, on the other hand, do not require the use of MHCs to detect cancer antigens, as they are able to recognize cell surface antigens.¹³ These receptors have a high-affinity recognition domain which activates downstream signaling pathways to direct immune cell cytotoxicity towards tumor cells.¹¹ Recent advancements in NK cell therapy include the development of CAR-NK cells, which are similar to CAR-T cells but offer a key advantage in that there is no need for human leukocyte antigen (HLA) matching. This makes allogenic transfers safer, and increases the likelihood that an “off-the-shelf” immunotherapy could be developed.¹⁴ CAR macrophages (CAR-Ms) have also been developed in an attempt to take advantage of the inherent recruitment of macrophages by the TME and the phagocytic abilities of macrophages. CAR-Ms expressed an anti-tumor phenotype, induced the same phenotype in bystander macrophages, and performed antigen-specific phagocytosis.¹⁵ Anti-tumor activity in immune cells can also be achieved via the use of cancer vaccines. Cancer vaccines can be cell-based or peptide/protein based. They either deliver immune cells that are activated to act against cancer or components that activate endogenous immune cells, including irradiated tumor cells and synthetic antigens.¹⁶ By combining the natural tumor targeting abilities of immune cells with bioengineering techniques, more effective immunotherapies can be developed.

Although immunotherapies have shown promise in exploiting immune cells to fight cancer, patient responses to treatment are still highly variable. Only a subset of patients see positive results with any given immunotherapy, and there is the potential for severe adverse effects.¹⁷ This demonstrates the critical need for effective monitoring of the immunotherapeutic process to improve current treatment workflows. Traditional methods of monitoring the progress of administered cell-based therapies rely on biopsies, which are invasive and do not provide accurate or timely information about in vivo events.^18,19 Additionally, biopsies examine only a small tissue sample. As a result, they are unable to provide spatiotemporal data about the number of activated immune cells or the location or viability of these cells. This makes it difficult to understand immune cell pharmacokinetics or predict therapeutic efficacy, and poses a particular challenge in cases of tumor heterogeneity.^18–21 Given the variability in outcomes of different cell therapies and the possibility of adverse effects, it is important to monitor these treatments, both to gain a better understanding of their mechanisms and to intervene if there is danger to the patient.^18,22,23

To address these shortcomings, researchers have developed novel methods for non-invasive in vivo cell monitoring. These methods use imaging modalities such as optical imaging, computed tomography (CT), magnetic resonance imaging (MRI), positron emission tomography (PET) and single-photon emission computed tomography (SPECT), and photoacoustic (PA) imaging. These imaging modalities can be used in conjunction with various contrast agents to obtain different kinds of information, from anatomical details to insights into molecular processes. This review compares the different methods used to noninvasively track cell-based therapies, highlighting the advantages and limitations of each approach. Here we provide a comprehensive overview of novel tracers and contrast agents being developed to track cell-based therapies in vivo that may not have been explored in detail elsewhere. This includes advancements in tracers that can reveal an overactive immune response that, if otherwise left untreated, could lead to graft-versus-host disease (GVHD), a reoccurring side effect of certain cell therapies.²² Our emphasis on the recent advancements in non-invasive imaging of immunotherapies gives innovative solutions to how we can further improve cell-based therapies. By exploring these cutting-edge imaging techniques, a more comprehensive understanding of cell-based therapies can be achieved, which not only enhances our ability to monitor therapeutic efficacy, but also helps us understand how to improve and optimize existing therapies (Fig. 1).

Fig. 1.

Representative image of the cell based therapies and imaging modalities used to monitor immunotherapies. Created in BioRender. A. Kulkarni (2024) https://BioRender.com/u57r022.

2. Cell-based immunotherapies

Developed in the 1960s, adoptive cell therapy is an innovative immunotherapy that repurposes a patient’s own immune cells to act as a cancer therapeutic.²⁴ Today, the field has expanded to include the use of a variety of immune cells such as T cells, NK cells, macrophages and others, with each cell type possessing unique properties that allow them to target cancer. Before exploring the techniques for monitoring cell-based immunotherapies, this section summarizes the different types of cell therapies and the need for immunotherapy tracking.

2.1. T cell therapies

Adoptive cell therapies using T cells can be divided into three main types: TIL therapy, TCR therapy, and CAR-T therapy. TIL therapy involves harvesting a patient’s T cells, often from a tumor biopsy, expanding them, and then reinfusing them back into the patient. This creates a larger population of T cells that can recognize and mount an effective immune response against tumor cells.²⁵ In clinical settings, patients are lymphodepleted prior to TIL to reduce the chance of adverse effects due to conflict with pre-existing T cell populations, such as GVHD.²⁶ Even though TIL therapy can be very effective, especially in combination with other therapeutic strategies, it can only be used when resectable tumors are present with enough T cells to harvest and expand.²⁵

To address this limitation, T cells can be genetically engineered with receptors that enable them to more effectively target tumor cells, such as in TCR therapy and CAR-T cell therapy. For TCR therapy, T cells are isolated from a patient’s peripheral blood and engineered to express tumor-antigen-specific T cell receptors, which have extracellular α and β chains that recognize cancer antigens presented by the MHC on the tumor cell surface.^18,27 These cells are activated by the interaction between the modified receptor and other signaling molecules such as CD3, which causes them to target tumor cells.²⁸ Complications can arise with TCR therapy when the α and β chains of exogenous TCRs crosslink with the α and β chains of endogenous TCRs. This crosslinking can create new TCRs that have entirely new dynamics for recognizing and responding to certain antigens, risking unintended immune reactions.²⁹

Similarly to TCR therapy, CAR-T cell therapy involves the genetic engineering of T cells, but specifically induced to express chimeric antigen receptors.²⁷ CARs recognize tumor-specific proteins on the surface of tumor cells, without the need for presentation by an MHC complex. This method allows for targeting of cancer cells at a greater scale than TCR therapy, as it circumvents the issue of MHC downregulation by cancer cells.¹³ Key components of CAR-T cells include an antigen-binding domain, an activation domain, and one or more costimulatory domains.³⁰ Several generations of CAR-T cells have been developed with augmented receptors. These receptors have additional costimulatory domains for greater specificity and cytotoxicity when activated by a certain antigen target.¹⁴ Despite the considerable success of CAR-T cell therapy for treatment of hematological malignancies, it remains ineffective at treating solid tumors due to unintended adverse effects of the treatment and the complexity of the TME.³⁰

2.2. NK cell therapy

NK cells are a part of the innate immune response and they exert cytotoxic activity against a target by producing cytokines.³¹ NK cells were proposed for use in cell therapy due to the sheer diversity in their cancer targeting mechanisms.¹⁴ Activated NK cells secrete cytolytic granules such as granzyme B and perforin to kill target cells via apoptosis.³² In addition, NK cells can use death receptors and pro-inflammatory cytokines to limit the growth of cancer cells and recruit other immune cells to the tumor site.³³

The first major development in NK cell therapy was autologous transfers, where patients received transfusions of their own NK cells. It was found that autologous NK cells were able to persist in patients’ blood up to several months post-transfer, though these NK cells had reduced tumor targeting capabilities.³⁴ In addition, patients receiving autologous NK cell infusions were often lymphodepleted prior to treatment, which can lower their NK cell counts and function.³⁵ Due to these limitations, there was a shift toward allogeneic NK cell transfers. Allogeneic NK cell transfers have been shown to have several advantages. This includes reduced toxicities compared to some T cell therapies, specifically lower occurrences of cytokine release syndrome (CRS) and GVHD. This therapy is also more universal than T cell therapy since NK cells target cancer cells in a HLA independent manner.³⁶ Additionally, allogenic transfers allow for the development of “off-the-shelf” NK cell therapy, which increases patient access to cell therapy.³⁷

These advances in NK cell therapy have led to the development of CAR-NK cell therapy. CAR-NK cells take advantage of the NK cell’s innate anti-tumor properties without the need for prior sensitization to tumor antigens.³⁸ CAR-NK cell therapy has several benefits over CAR-T cell therapy, including a reduced risk for GVHD and CRS since NK cells have a higher activation threshold and do not require antigen-specific activation, resulting in a more controlled cytokine release.^38,39 Even though there have been significant advances in CAR-NK cell therapy, there are still many limitations. For instance, the therapeutic efficacy of this treatment in solid tumors is low, which has to do with NK cells interactions in the TME. The TME is inhospitable to NK cells due to poor physiological conditions, disruption of NK cell signaling, secretion of tumor cell-derived factors and the activity of immunosuppressive cells. Hypoxia in solid tumors has been shown to inhibit NK cell anti-tumor activity by preventing NK cell recognition of tumor cells. In addition, regulatory T cells (Tregs) and myeloid-derived suppressor cells had the effect of inhibiting NK cell activity and proliferation in the TME.^40,41 Further research must be done in order to increase the efficacy of NK cell therapies. Potential strategies include improving NK cell targeting abilities, reducing inhibitory signaling pathways, and enhancing NK cell persistence.³⁷ Implementation of these strategies requires a way to monitor NK cell therapy in vivo.

2.3. Macrophage and monocyte therapies

Macrophages are white blood cells that act to clear damaged cells or foreign bodies via the process of phagocytosis, as well as stimulate other immune cells.^42,43 Macrophages in the TME can exhibit either pro-tumor or anti-tumor activity. Tumor-associated macrophages (TAMs) are programmed by the TME to support tumor growth and metastasis (M2 phenotype) by promoting cancer cell functions, suppressing immune responses and releasing cytokines that promote a metastatic niche. However, macrophages can be reprogrammed to kill tumor cells (M1 phenotype) or can be the target of therapies themselves.⁸

With the goal of reprogramming macrophages, CAR-Ms have been genetically engineered to induce phagocytic activity against tumor cells. The introduction of the viral vector used to engineer CAR-Ms induces an M1 phenotype, regardless of the gene that is being delivered. Expression of CARs by CAR-Ms enables direction of phagocytosis towards cancer cells. Furthermore, CAR-Ms express cytokines that induce bystander macrophages to an M1 phenotype and increase antitumor T cell activity, resulting in decreased tumor burden.¹⁵ TAM polarization can also be reverted or induced to an M1 phenotype via the use of anticancer drugs such as gemcitabine and 5-fluorouracil.^44,45 Alternatively, macrophages can be the target of immunotherapies. Germano et al. found that the chemotherapeutic agent trabectedin exhibited specific cytotoxicity against TAMs via the activation of caspase-8 and caspase-9, which play a role in the apoptotic pathway. This results in reduced pro-tumor activity by the macrophages, which could result in better patient prognosis.⁴⁶

Monocytes are precursors to macrophages and dendritic cells, and they are also found in the tumor microenvironment. Much like macrophages, monocytes in the TME have a dual role, in that they can promote tumorigenesis and metastasis, but can also act to inhibit metastasis. Monocytes have been used in immunotherapies by reprogramming them to an inflammatory phenotype that kills cancer cells.⁴⁷ Specifically, it was determined that artesunate treatment could be used to program monocytes to kill leukemia cells.⁴⁸ Another strategy involving monocytes for cancer therapeutics is to prevent their recruitment to tumor sites so that they cannot contribute to tumorigenesis and metastasis.⁴⁷ Losartan has been used to disrupt the CCL2–CCR2 chemotactic axis, which is the driving force for monocyte recruitment for metastasis is lung cancer. This treatment greatly reduces monocyte recruitment and results in lower metastatic burden.⁴⁹ Monocytes can also be used in immunotherapies as a drug delivery vehicle. They naturally accumulate at tumor sites, making them ideal for delivering drug-loaded nanoparticles or particles for photodynamic therapy.^9,47,50

While these therapies show promise in the treatment of cancer, it is still vital that the cells be tracked following injection in order to further immunotherapy development and understand therapeutic outcomes.⁵¹

2.4. Dendritic cell therapy

Dendritic cells (DCs) are phagocytic antigen presenting cells that play a vital role in activating T cells in the immune system. DCs recognize foreign antigens, which triggers their maturation and causes them to engulf the corresponding pathogen. They present these foreign antigens to T-cells at the lymph nodes, inducing an immune response.^52,53 Numerous immunotherapies have been developed utilizing DCs due to the vital role they play in activating the immune response.¹⁰

The immunotherapy strategies that have been developed using DCs include in vivo activation, in vivo expansion, blocking of inhibitory signals, whole cell vaccines and peptide/protein vaccines (vaccine strategies are discussed in depth in the following section). In vivo activation therapies attempt to counteract the suppression of DCs induced by the TME. These treatments use exogenous activation signals such as toll-like receptor agonists to induce maturation and anti-tumor responses in DCs.^10,53,54 Another immunotherapy strategy is to increase the number of DCs present in the TME via in vivo expansion.¹⁰ Within tumors there are very few DCs present, which limits their therapeutic potential.⁵⁵ By injecting growth factors, DC populations within tumors can be expanded and activation is induced, which leads to improved tumor control.⁵⁶ This strategy has advantages in that patient specificity is unnecessary and a wide range of antigens is targeted as opposed to the limited antigens targeted by other therapies.¹⁰ Signals from the TME that inhibit DC function can also be blocked. Vascular endothelial growth factor (VEGF) is common in the TME and acts to inhibit DC function. VEGF inhibitors can be administered systematically to increase antitumor activity by preventing DC inhibition and angiogenesis.^10,57 While these therapies are promising, their therapeutic efficacy is lower than expected, which is in part due to inaccuracies with DC therapy injections, insufficient lymph node homing and improper timing of treatment. The ability to track DCs following administration of therapy would increase understanding of DC behavior and allow for further development.⁵⁸

2.5. Cancer vaccines

Both preventative and therapeutic vaccines have been developed for the management and treatment of cancer.¹⁶ The first category of cancer vaccines is cell vaccines, which includes autologous tumor cell vaccines and dendritic cell vaccines. Autologous tumor cell vaccines use irradiated tumor cells and an adjuvant to trigger immune responses that result in antitumor activity by endogenous immune cells.¹⁶ The use of whole DCs in a cancer vaccine is a newer therapy, developed to take advantage of the central role DCs play in the immune system, as previously mentioned. DCs can be stimulated ex vivo with different antigens to create a personalized treatment depending on the condition of the patient, which can include taking patient samples to create the vaccine. Furthermore, DC vaccines have low toxicity, making them a safer approach than other therapies. However, efficacy of this treatment has been limited thus far.^10,16,59

Protein/peptide vaccines use a similar mechanism, though they influence cells in vivo rather than ex vivo. Synthesized antigens are introduced and taken up by antigen presenting cells such as DCs, which leads to an immune response when T cells recognize the antigen. However, most clinical trials using protein/peptide vaccines fail in later phases, even after initially showing success.^16,60 The limited success of cancer vaccines despite their ability to induce an immune response demonstrates the need for immune cell tracking, as it would elucidate the reasons for failure and allow for improvement of the therapy.

2.6. Stem cell therapy

Stem cells have been a contender for use in cell based therapies due to their ability to home to the TME and target some cancers directly. Various kinds of stem cell therapies have been developed for the treatment of cancer. Of particular interest are bone marrow derived stem cells, including mesenchymal stem cells (MSCs) and hematopoietic stem cells (HSCs). MSCs are used in cell therapy applications due to their ability to regenerate connective tissues, including bone, cartilage and adipose tissues. They also home to areas with tissue damage.^61,62 Various animal models have demonstrated the antitumor properties of MSCs, leading to reduced growth of glioma, melanoma, lung cancer, hepatoma, and breast cancer cells.⁶² Hematopoietic stem cell transplantation (HSCT) is a type of cell therapy used for a range of diseases, including hematological malignancies.²² However, in some instances, HSCT therapies have led to complications related to GVHD. It has been shown that HSCT creates alloreactive donor T cells that attack host cells, resulting in GVHD. This is of particular concern since diagnosis for GVHD typically occurs once the disease has already progressed.²² The occurrence of GVHD highlights the urgent need for improved monitoring and management strategies in cell therapies to mitigate these risks and enhance patient outcomes.

Stem cell therapies have a wide range of applications, including immunotherapies for cancer. While these therapies show promise as treatments for some malignancies, their use has been limited due to their potential for both pro-tumor and anti-tumor effects observed in preclinical studies. This highlights the need to track these therapies and assess their function in vivo.⁶³

2.7. The need for immune cell tracking

Each type of immunotherapy has distinct advantages based on the immune cell and bioengineering strategy that are utilized. Though these immunotherapies have been demonstrated to be effective in the treatment of cancers, they each have limitations. In order to increase the efficacy of these treatments and improve patient prognosis, these limitations must be addressed. This can be accomplished by developing techniques to track the cells used in immunotherapies. Tracking cell based therapies can provide critical information about the localization, activation and viability of immune cells, furthering development of immunotherapies and allowing for prediction of therapeutic efficacy.²⁷ Traditionally, tracking of immunotherapies have occurred with biopsies, which are invasive, provide data at only one time point and are limited by sampling errors such as tissue morbidities.⁴ To address the lack of real-time, non-invasive monitoring, various techniques have been developed to track immune cells using different imaging modalities.

3. Monitoring immune cells using different imaging modalities

Non-invasive imaging of immunotherapies enables providers to monitor the real time presence, distribution, and dynamics of the immune cells used in cell therapies (Table 1).²³ Different methods of tracking are employed depending on the imaging modality being used. Cells do not naturally possess any reporter molecules to generate contrast that can be imaged, so the contrast must be introduced to the cell. These methods can be generalized into either a materials engineering approach or a biological engineering approach. For materials engineering, cells are labeled with exogenous contrast agents. The contrast agents are internalized in the immune cells, which allows them to be tracked. For the biological engineering approach, cells are transduced to express reporter genes which bind to a particular tracer molecule and can be used for visualizing the cell. These in vivo tracking methods offer real-time insights that can be used to evaluate the effectiveness of a treatment without requiring invasive procedures (Table 2).

Table 1.

Overview of different in vivo imaging modalities along with their pros and cons

Imaging modality		Pros	Cons
Optical imaging	BLI	- High sensitivity - High signal to noise ratio - Low background signal in small animal models	- Limited imaging depth - Non-mammalian origin of luciferase causes immunogenicity issues in large animal models
	FLI	- Low cost - Can track multiple different probes in vivo simultaneously	- Limited imaging depth - Poor spatial resolution
	IVM	- High resolution - Understand interaction of cells in TME	- Limited imaging depth
	OCT	- High resolution - Fast scan times - 3D images	- Limited imaging depth - High cost
X-ray/CT		- No depth limitation - Anatomical information - Whole body imaging - Faster scan times compared to MRI	- Exposure to ionizing radiation - Lower soft tissue contrast compared to MRI
PET		- No depth limitation - High sensitivity - Whole body imaging	- High cost - Exposure to ionizing radiation - Lower accessibility than SPECT
SPECT		- No depth limitation - Whole body imaging - Greater accessibility than PET	- Exposure to ionizing radiation - Lower sensitivity than PET
MRI		- No depth limitation - Anatomical information - Whole body imaging - High resolution - Better soft tissue contrast compared to CT	- High cost - Lower sensitivity than PET/SPECT - Long acquisition times
US/PA		- High spatial resolution - Multiplexing capabilities - Anatomical information from ultrasound	- Limited penetration depth

Table 2.

Current immune cell tracking methods for immunotherapy, including contrast agent, cell therapy and imaging modality

Imaging modality	Nanoparticle/contrast agent	Cell-based therapy	Application	Ref.
BLI	Firefly luciferin	NK cell therapy	Tracking immune cell migration, gene expression, and tumor growth	64 and 65
FLI	Fluorescent quantum dots (QD705)	NK cell therapy	Imaging immune cell distribution and activity, monitoring therapeutic efficacy	64 and 66
FLI	1,1-Dioctadecyltetramethyl indotricarbocyanine iodide (DiR)	Adoptive immune cell therapy with T-lymphocytes	Image and assess tumor targeting and retention of directly labeled T-lymphocytes	67
IVI	tdTomato fluorescent protein	CAR-T cell therapy	Monitor in vivo antitumor reaction of CAR-T cells on PCNSL	68
CT	Bi2S3 Nanoparticles modified with ovalbumin	Dendritic cell-based cancer vaccine	Monitor DC migration to lymph nodes and induce anti-ovalbumin immune activity	69
CT	20 nm GNPs	TCR Therapy	Tracking T-cells accumulation at tumor site to determine efficacy of TCR therapy for melanoma	70
CT	20 nm GNPs	NK cell therapy	Longitudinally tracking NK cell accumulation at the tumor site	71
CT	Dendrimer entrapped GNPs	Macrophage therapy	Macrophage polarization and tracking in vivo	51
CT	GNPs modified with 11-mercaptoundecanoic acid	Monocyte therapy	Tracking monocyte accumulation in atherosclerotic plaques	72
CT	80 nm GNPs	Mesenchymal stem cell therapy	Tracking mesenchymal stem cell migration in the subretinal layer	73
PET/SPECT	111Indium (In)-tropolone	CAR-T cell therapy	Monitor T cell trafficking in vivo. Engineering cells using radiotracers.	74
PET/SPECT	[18F]F-AraG	Stem cell therapy	Track alloreactive T cells, indicate acute GVHD.	22
PET/SPECT	89Zr-oxine	CAR-T cell therapy	Track CAR-T cells for up to 6 days post transfer, compare administration techniques.	75
PET/SPECT	[18F]fluoropropyl-trimethoprim	CAR-T cell therapy	Tumor monitoring and homing	76
MRI + FLI	PLGA nanoparticle containing SPIO + ICG + ovalbumin	Dendritic cell-based cancer vaccine	DC activation and monitoring migration via FLI and MRI	77
MRI	19F PFPE	Dendritic cell therapy	In vivo tracking of dendritic cell migration	78
MRI	19F PFPE	Adoptive NK cell therapy	Longitudinal tracking of NK cells following intratumoral injection	79
MRI	19F PFC	Splenocyte and T cell therapy	Whole body monitoring of splenocytes and T-cell behavior and migration	80
PA	NIR-797-isothicyanate	T cell therapy	Monitoring T cell migration to tumor site
PA	Semiconducting polymer nanoparticle	Macrophage therapy, neutrophil therapy and MSC therapy	Monitoring macrophage, neutrophil and MSC migration to inflammation site	81
PA	Gold nanorods	Dendritic cell therapy	DC activation and longitudinal tracking of migration to lymph nodes	82
PA + FLI	Gold nanorods coated in fluorescent dye doped silica	Macrophage therapy	In vitro bimodal imaging of macrophages	83
US/PA	Silica coated gold nanorods	T cell therapy	In vivo tracking of T cells in a solid tumor and monitoring oxygen saturation	84
PA	Silica coated gold nanorods	Cytokine induced killer cell therapy	Imaging of CIK cells, enhanced antitumor activity and photothermal therapy for gastric cancer	85
PA	Gold nanoprisms	Cytokine induced killer cell therapy	Imaging of CIK cells, enhanced antitumor activity and photothermal therapy for gastric cancer	86

3.1. Optical imaging

Optical imaging uses light to visualize and monitor biological processes. This type of imaging is widely used preclinically, as it is a cost-effective and high-resolution method.⁶⁴ One optical imaging technique is bioluminescence imaging (BLI), which involves transfection of cells to express a luciferase enzyme that converts molecules called luciferins into light.^4,87 The luciferase enzyme comes from one of several bioluminescent organisms. Luciferins have to be added to cells expressing this enzyme for imaging, as it is used in a reaction catalyzed by luciferase along with oxygen, ATP, and cofactors to produce light (Fig. 2b).^23,27,64,87 Light emissions from luciferins range from 530 nm to 640 nm, which allows for several centimeters of imaging depth in biological tissue.⁸⁸ This method of cell labeling using BLI in small animal models allows for tumor cells to be seen at high sensitivity since mammals do not express luciferase endogenously, giving it a high signal-to-noise ratio and low background signal.⁶⁴ However, the non-mammalian origin of luciferase is also a drawback with BLI, as it cannot be applied in humans due to its immunogenicity.^4,27 BLI is primarily used to track tumor growth and migration, but it can also be used with immune cells that have been modified to express luciferase in order to track immune cell migration.^64,65 In a study by Zhu et al., human NK-92 cells and anaplastic thyroid cancer (ATC) cells were transduced with different luciferase genes in order to perform long-term monitoring of the therapeutic effects of NK cells in vivo against ATC.⁶⁵ ATC is an aggressive solid tumor malignancy which is unresponsive to conventional chemotherapy or radiotherapy.⁸⁹ Thus, cell-based immunotherapy approaches have become an important area of investigation against ATC. The use of two different luciferase-based reporter genes allowed for BLI imaging of both NK cells and ATC cells simultaneously. The effect of NK cells on ATC metastases in vivo was tested using the BLI signal intensity to assess if the NK cells were effective. Seven days after NK treatment, the BLI signal intensity was 2.8-fold higher in the control group than in the NK treatment group, indicating that the tumor shrank in the mice treated with NK cells and demonstrating NK cell cytotoxicity against ATC cells.⁶⁵

Fig. 2.

Optical imaging for in vivo immune cell tracking. (A) Fluorescence comprises the excitation of a fluorescent molecule, causing its electrons to move to a higher energy state, then return to a ground state, releasing the energy as light. This light is imaged with FLI.⁶⁴ (B) Bioluminescence occurs when a luciferase enzyme catalyzes the reaction of a luciferin with oxygen, ATP, and cofactors to produce light that can be measured with BLI.⁶⁴ Copyright © The Author(s). Cancer Medicine published by John Wiley & Sons Ltd. (C) Bioluminescence imaging of a tumor expressing luciferase. In the control, the tumor continued growing over 7 days, causing the signal to spread and become brighter. In mice treated with NK cells, the signal decreases as the tumor shrank from the treatment.⁶⁵ Copyright © 2017 Zhu, Li, Kalimuthu, Gangadaran, Lee, Oh, Baek, Jeong, Lee, Lee and Ahn. (D) Fluorescence imaging to observe nitric oxide produced by macrophages treated with different immunotherapy drugs responding to a tumor in a mouse model.⁹⁰ Copyright © The Author(s). Advanced Materials published by John Wiley & Sons Ltd. (E) Mice were injected with DiR labeled T cells one week prior to introducing 4T1, a murine breast cancer cell line, then monitored with fluorescence imaging after 4T1 injection. T cells migrated to the tumor within 2 hours and the signal remained mostly stable for at least 2 weeks.⁶⁷

A second method of optical imaging is fluorescence imaging (FLI). FLI requires the use of fluorescent molecules, such as near-infrared dyes, that emit a specific wavelength of light upon excitation from an external light source (Fig. 2a).^27,67,91 Optical imaging methods are limited by tissue attenuation of light, however, advances in the engineering of near-infrared fluorescent probes have enhanced tissue penetration as near-infrared light undergoes less attenuation.⁶⁴ These probes can be designed to be lipophilic to label the cell membrane, such as 1,1-dioctadecyltetramethyl indotricarbocyanine Iodide (DiR), a NIR fluorescent dye.⁶⁷ A variety of NIR probes exist, each with a unique spectral signature, allowing for imaging of two or more different types of fluorophore-labeled cells at once to assess therapeutic efficacy.^64,91 This enables assessment of immune cells penetration into tumors by labeling the two cell types with different fluorophores and unmixing the signals following imaging.⁹¹ FLI has been used to track numerous cell-based therapies, including T cell therapies. A study by Sauer et al. demonstrated the effectiveness of adoptively transferred T cells against acute myelogenous leukemia (AML), a liquid tumor often treated with multiagent chemotherapy.⁹² In older adults with AML, however, chemotherapy has led to significant risk of developing toxicities which lower the likelihood of remission.⁹³ Thus, alternative strategies involve the use of FLI to track adoptively transferred cytotoxic T cells (CTLs) targeting AML. To simultaneously track two different cell populations Sauer et al. created transgenic murine AML cell lines expressing DSRed2, a red fluorescent protein, along with transgenic anti-AML CTLs expressing enhanced green fluorescent protein (eGFP), which were detected using fluorescence imaging.⁹² Longitudinal tracking of CTL and AML cells using FLI allowed investigators to understand the migration patterns of CTLs that could be important for an effective treatment. This includes the finding that no significant trafficking of CTLs were found in the central nervous system (CNS), which may be a major area of improvement for the therapy. With FLI it was possible to understand potential pitfalls of CTL therapy, which will allow investigators to optimize strategies for enhancing the efficacy of these treatments in the long run.

In addition to T cells, FLI has been used to inform the effectiveness of NK cell therapies. A study by Lim et al. tracked NK cells labeled with QD705, a quantum dot (QD) that absorbs NIR light.^64,66 Lim et al. used QDs coated in antihuman CD56 antibodies to allow them to bind to CD56 receptors on the NK cells, circumventing the issue of limited QD internalization by NK cells. By measuring and comparing interferon gamma production and cytolytic activity of labeled and unlabeled cells, they were able to prove that fluorescently labeled NK cells were equally effective as unlabeled NK cells for cancer therapy. This method of labelling and imaging NK cells could be implemented in preclinical trials to determine if NK cell therapy is effective against different types of cancer.⁶⁶

Intravital microscopy (IVM) is another method of optical imaging that uses light microscopy to observe cells in vivo through a window implanted into an animal’s body.⁶⁴ This method allows for high resolution imaging of the interactions between cells in the tumor microenvironment. There are different techniques of IVM, including intravital multiphoton microscopy, which uses two or more photons in the near-infrared wavelength to excite fluorophores that can be seen under a microscope.⁹⁴ This method has been implemented to monitor the effectiveness of CAR-T cell therapy for treatment of primary central nervous system lymphoma (PCNSL), which is a solid tumor.⁶⁸ Previous CAR-T cell therapies have shown limited successes in solid tumors due to decreased migration of T-cells into the tumor and the immunosuppressive TME.⁹⁵ In this study, CAR-T cells and cancer cells were transfected to express different fluorescent proteins, including tdTomato and eGFP, that could be imaged in vivo using two-photon microscopy. Using IVM, researchers were able to analyze the distribution of CAR-T cells, with signals persisting for 159 days. The use of IVM was therefore able to show CAR-T cells trafficking into the PCNSL solid tumor and regression of the malignancy.⁶⁸

Optical coherence tomography (OCT) is an additional optical imaging technique that utilizes low-coherence interferometry to generate two- or three-dimensional images of tissue.^96,97 However, the limited imaging depth of OCT limits its capabilities for in vivo imaging of non-transparent tissues. Currently, OCT is only capable of imaging up to 2 mm deep, which limits its potential as a non-invasive imaging method to only tumors on the skin’s surface.⁹⁸ While OCT has been employed for cancer diagnosis and monitoring, its application has primarily been through invasive procedures such as endoscopy and biopsy.^99,100 To date, no research has explored the use of OCT for in vivo tracking of immune cells; however, advancements in OCT technology aimed at improving penetration depth could enhance its potential as a non-invasive tool for monitoring cell-based immunotherapies.

Overall, optical imaging is a useful tool to monitor the effects of immunotherapy in the TME. However, a major limitation of optical imaging is tissue attenuation of light, which limits imaging depth. As a result, this method can only be used for in vivo imaging of small animal models.^4,27,64 Despite this limitation, it is still widely used in preclinical trials since it allows for real-time, dynamic imaging of the response to cell based immunotherapies.

3.2. X-Ray computed tomography

While optical imaging does show some promise for in vivo tracking of immune cells, its limited tissue penetration depth and low spatial resolution pose significant challenges for clinical translation.²³ Imaging modalities such as computed tomography (CT) address these issues, in addition to providing information about internal anatomy.¹⁰¹ CT can be used with or without contrasts depending on the objective. In cases where tumors are well visualized and size and shrinkage are indicative of treatment progression, contrasts are not necessary and anatomical imaging (MRI, CT) is sufficient.^4,102 However, tumor size is not always indicative of the efficacy of treatment, as is seen in cases of pseudoprogression and pseudoresponse, where tumor size increases or decreases in a way that does not reflect tumor burden. In such cases, anatomical imaging will not give accurate information on treatment progression and methods that track immune cells or tumor cells directly are necessary.^4,103,104

CT can be combined with contrast agents to achieve molecular imaging, which enables tracking of molecular processes throughout the body, including monitoring the efficacy of immunotherapies.⁴ Contrast agents that directly interface with various immune cells have been developed as a result. CT contrast agents can be composed of a variety of different materials, including lipids, iodine, bismuth, lanthanide elements, and transition metal elements such as gold.^105,106 Bismuth based nanoparticles have been used to track the activity of DCs in a cancer vaccine formulation while simultaneously acting as an antigen carrier. Bi₂S₃ nanoparticles were biomineralized with ovalbumin and then loaded into DCs. The nanoparticles were found to increase immune activity by inducing DC maturation and antigen presentation, as well as promoting anti-ovalbumin activity by T cells. Furthermore, the migration of DCs to the lymph nodes was able to be tracked using CT imaging due to the high X-ray absorption rate of bismuth.⁶⁹

While there have been some studies into contrast agents composed of other materials, gold-based contrast agents are by far the most popular for cell tracking using CT imaging. Gold nanoparticles (GNPs) have proven to be advantageous due to their high X-ray attenuation properties, biocompatibility, and the ease of synthesis and surface modification (Fig. 3a). The use of GNPs allows for better visualization of tissues and cells, as well as lowering the radiation dosage needed for visualization.^106,107

Fig. 3.

X-ray computed tomography for in vivo immune cell tracking. (A) Cell tracking procedure for CT imaging. Cells are labeled with GNPs in vitro and are then injected intravenously into mice and tracked using CT imaging.⁷¹ Copyright © 2021 The Authors. Published by American Chemical Society. TEM images of (B) control macrophages without treatment and (C) macrophages treated with dendrimer-entrapped gold nanoparticles.⁵¹ Copyright © 2021 The Authors. Chemical Engineering Journal published by Elsevier. (D) 3D CT, CT and fluorescence images of tumor. 3D CT image indicates the location of the tumor, while CT and fluorescence track GNP labeled T-cells used in TCR therapy. There is a high correlation between CT imaging and fluorescence imaging, validating the tracking of T-cells using GNPs and CT.⁷⁰ Reprinted (adapted) with permission from R. Meir, et al., ACS Nano, 2015, 9(6), 6363–6372. Copyright 2015 American Chemical Society. (E) 3D CT image of GNP labeled NK cells accumulating at the tumor site following intravenous injection of NK cells.⁷¹ Copyright © 2021 The Authors. Published by American Chemical Society.

GNPs were used to label T-cells that were being used for the treatment of melanoma. The study by Meir et al. demonstrated the use of 20 nm GNPs surface modified with polyethylene glycol for stability and glucose to increase cellular uptake. The T-cells were modified to express a melanoma-specific TCR, which led to increased accumulation of T-cells at the tumor site in mice following intravenous injection. T-cells were then labeled with GNPs in vitro, and GNPs were found to have no effect on T-cell function or viability, as evidenced by tumor regression. There was a strong correlation between signal intensity of GFP tagged T-cells tracked with fluorescence imaging and CT signal, demonstrating the accuracy of cell tracking (Fig. 3d).⁷⁰ This research group also successfully labeled NK cells with GNPs using the same method, and showed increasing CT signal up to 72 hours (Fig. 3e).⁷¹ GNPs have also been used as a theranostic with TAMs. Macrophages were engineered with dendrimer-entrapped gold nanoparticles for treatment of osteosarcoma (Fig. 3b and c).⁵¹ Dendrimers increase stability of the GNPs in a wide range of environmental conditions, as well as enabling modification of the nanoparticle surface with drugs or targeting ligands.¹⁰⁸ In this case, labeling with GNPs encouraged the macrophages to an M1 phenotype, which has antitumor effects. Thus, these nanoparticles act as both an imaging contrast and as a therapeutic for treating osteosarcoma.⁵¹

Other immune cells have also been labeled for CT imaging, though it was not specifically for use in cancer treatment. Despite this, the successful labeling of these immune cells suggests that these contrasts could be used for cell tracking with immunotherapies. Monocytes were labeled with GNPs modified with 11-mercaptoundecanoic acid as a stabilizing agent, and monocyte function and viability were unaffected. Increased contrast was seen with CT in mice injected with the labeled cells, and monocyte accumulation was seen within the atherosclerotic plaques that were the target of this treatment.⁷² Longitudinal tracking of mesenchymal stem cells was accomplished using 80 nm GNPs for repair of the subretinal layer in rats. Signal was detected up to 30 days, which suggests that this could be a promising treatment and longitudinal cell tracking method for cancers of the eye.⁷³ More work must be done to validate CT imaging contrasts for the specific purpose of immune cell labeling for cancer immunotherapies, including contrast agents other than GNPs. While GNPs have proven to be an effective method of monitoring immune cells thus far, there are concerns that the high cost of gold could limit its clinical translation, an issue that could be addressed by further study into other materials.¹⁰⁵ The development of contrasts that reduce the radiation dosage would also further the use of CT for immune cell tracking.

3.3. Positron emission tomography (PET) & single-photon emission tomography (SPECT)

Radionuclide imaging, such as positron emission tomography (PET) and single-photon emission computed tomography (SPECT) can also be used to monitor the behavior of immune cells. These techniques rely on the use of radiotracers, whose emissions are detected by specialized nuclear imaging cameras. When combined with CT in hybrid systems, the anatomical details provided by CT can improve localization and interpretation of the functional data. PET and SPECT use different radiotracers that have varying levels of sensitivity to detect abnormal tissues and tumors. PET tracers emit positrons that react with electrons in the body to form photons that are detected by PET scanners to generate images. With SPECT, gamma rays are released from the radionuclides and detected using a gamma camera, which can then be used to create 3D reconstructions of the signal.¹⁰⁹ Compared to SPECT imaging, PET imaging generally provides higher sensitivity and improved spatial and temporal resolution for visualizing biological processes.¹¹⁰ While SPECT is more accessible and cost-effective in many clinical settings, it is employed less frequently for advanced research applications than PET imaging, largely due to its comparatively lower sensitivity and resolution.¹¹¹ Regarding contrast agent development, certain PET isotopes occur naturally, a factor that can be both advantageous and disadvantageous. On one hand, leveraging naturally occurring isotopes may facilitate the synthesis of PET tracers with improved biocompatibility, yet these tracers may exhibit lower targeting specificity compared to SPECT tracers that employ heavier isotopes.¹¹² Thus, PET and SPECT each possess unique characteristics, providing distinct advantages for tracking cell-based immunotherapies.

Multiple methods of tracking in PET/SPECT imaging have emerged in immunotherapy, including material engineering of cells with exogenous agents such as radioactive tracers and biological engineering using reporter genes. Material engineering of cells for PET/SPECT imaging involves the use of different radionuclide tracers with varying properties. For example, Parente-Pereira et al. demonstrated the labeling with ¹¹¹Indium (In)-tropolone for the tracking of CAR-T cell therapy using SPECT imaging. The signal from ¹¹¹Indium (In)-tropolone can be detected for at least 96 hours after adoptive transfer. Using SPECT, T cell migration from the lungs to the liver, spleen and then lymph nodes could be tracked in immunocompromised mice. Imaging of tumor-bearing mice revealed lower levels of CAR-T cell trafficking to the tumor site compared to the liver when the treatment was administered systemically.⁷⁴ These findings put into question the tumor homing abilities of the cells. It is important to note, however, the trafficking patterns found in this study may be subject to the sensitivity of SPECT imaging. In fact, a different study tracking CAR-T cells using BLI did not see the same localization patterns of the T cell populations.¹¹³ The inconsistency between these two methods raises concerns about the ability of SPECT to provide accurate information about cell localization.

Recent investigations into novel PET tracers led to the development of the [¹⁸F]F-AraG radiotracer.²² The AraG compound is an essential component of this tracer, as it can inhibit DNA synthesis and induce cytotoxic effects towards T cells. In this way, the tracer targets over activated T cells. This is particularly useful when considering acute graft-versus-tumor (aGVHD) effects arising as a result of HSCT. Using [¹⁸F]F-AraG along with PET imaging techniques, overly activated T cell populations which cause this condition can be visualized, enabling early diagnosis of aGVHD. The study found that it was possible to track the progressive increase in alloreactive T cell populations in the lymphoid organs of aGVHD mice, providing a novel technique to track the effects of stem cell therapy.²²

Some studies demonstrated the use of a PET tracer, ⁸⁹Zroxine, to understand CAR-T cell localization with different administration techniques.⁷⁵ Due to the long half-life of ⁸⁹Zroxine, the study was able to track the treatment for up to six days post adoptive transfer.¹¹⁴ PET imaging showed that CAR-T cells traveled to the liver and spleen when administered intraventricularly, while intratumoral administration led to CAR-T cells remaining at the tumor site.⁷⁵ The ability to track cell therapies longitudinally provided invaluable information about the tumor homing and infiltration of CAR-T cell therapy. By engineering CAR-T cells with radioactive tracers, it is possible to visualize and compare administration techniques and improve upon cell-based immunotherapy treatments. However, engineering cells with exogenous contrast agents poses several challenges, including potential toxicity to immune cells from radiotracer labeling, signal reduction over time due to cell division, and limited longitudinal tracking capabilities arising from the short half-lives of many radiotracers.¹¹⁵

Biological Engineering approaches using reporter gene-based probes are an alternative to engineering cells with radiotracers. Biological engineering methods have several advantageous properties that make them ideal for in vivo longitudinal tracking for PET/SPECT. First, a reporter gene will only be expressed by live cells, which allows investigators to track the viability of the cells used in immunotherapies.¹¹⁶ In addition, for CAR-T cell therapy, it has been found that reporter genes do not reduce cell viability or antitumor activity.¹¹⁷ Lastly, there is no issue of signal loss due to cell division unlike other methods of labeling.¹¹⁴ A recent study by Sellmyer et al. utilized the reporter gene Escherichia coli dihydrofolate reductase (eDHFR) paired with a radiolabeled tracer, [¹⁸F]fluoropropyl-trimethoprim ([¹⁸F]-TMP). In this study, CAR-T cells expressing eDHFR targeted the GD2 glycolipid, an antigen presented by many cancers. These CAR-T cells were injected into mice bearing GD2+ and GD2− tumors. A significant amount of CAR-T cells were observed localizing to the antigen presenting tumor using PET/CT as compared to the antigen negative tumor (Fig. 4). It is important to note the high sensitivity of detection of the CAR-T cells infiltrating to the target area using the PET radiotracer, [¹⁸F]-TMP.⁷⁶ These PET images revealed the localization of the CAR-T cell therapy, allowing investigators to understand the homing and effectiveness of the treatment. Despite the advances in PET/SPECT imaging using reporter genes, limitations persist. To use reporter genes in a clinical setting, researchers need to create and receive approval for new types of CAR-T cell.¹¹⁰ In addition, the possibility of triggering an immune response against the reporter gene cannot be ruled out.¹¹⁸

Fig. 4.

PET computed tomography for in vivo immune cell tracking. PET/CT images of [¹⁸F]-TMP uptake in eDHFR and control tumor cells in vivo. High amounts of signal was detected in eDHFR cells compared to the control tumor, demonstrating CAR-T cell targeting abilities.⁷⁶ © 2019 The American Society of Gene and Cell Therapy.

While PET/SPECT imaging represents a significant advancement in the noninvasive tracking of cell based therapies, the associated challenges must be carefully navigated. PET/SPECT imaging is oftentimes costly and thus poses challenges with accessibility. Additionally, problems with low resolution can arise when CT is not being used, and radiotracers subject patients to radiation doses.¹¹⁹ As researchers continue to explore the potential of these imaging techniques, addressing these limitations will be crucial for ensuring their safe and effective integration into clinical practice.

3.4. Magnetic resonance imaging

MRI offers several advantages similar to those provided by CT and PET/SPECT, while additionally avoiding ionizing radiation and providing superior soft tissue contrast. These advantages have led to MRI being one of the more popular imaging modalities for immunotherapy monitoring, and many contrasts have been developed as a result.¹⁰²

One such example of these contrast agents are superpara-magnetic iron oxide (SPIO) particles. The core of SPIO particles is iron atoms, which leads to hypointensities and allows the particles to act as a negative contrast (Fig. 5a).^120,121 These particles can be encapsulated in liposomes or biocompatible coatings such as poly(lactic-co-glycolic acid) (PLGA), or conjugated with antibodies or antigens to increase selective labeling or prime immune cells respectively.⁵⁸ SPIO particles have been used in conjunction with peptide-based cancer vaccines to track both the components of the vaccine and the response of CD8+ cytotoxic T cells, Tregs and myeloid-derived suppressor cells to the vaccine.^101,122 They have also been used to track the recruitment of cytotoxic T lymphocytes and myeloid lineage cells to the tumor site and lymph nodes in a cervical cancer model in response to the combination of the DPX immunotherapy vaccine and anti-PD-1 therapy. Cells were tracked up till 28 days, and time-dependent differences in immune cell response to treatment were able to be monitored (Fig. 5b).¹²³ They have also been used to track dendritic cells for dendritic cell-based cancer vaccinations to enable verification of the correct injection site, adequate dendritic cell activity and the presence of adequate antigens for T-cell activation.⁵⁸ Specifically, DCs have been labeled with SPIO particles combined with protamine sulfate, which served to increase cellular uptake. DC function was maintained, and the migration of DCs to lymph nodes for antigen presentation to T cells was tracked.¹²⁴ SPIO particles have also been used for bimodal imaging and as a therapeutic, by encasing the particles with near-infrared fluorophores and the antigen ovalbumin in a PLGA nanoparticle. This enabled researchers to track DC migration through MRI and near-infrared fluorescence imaging, while simultaneously inducing an antigen specific immune response.⁷⁷

Fig. 5.

MRI for in vivo immune cell tracking. (A) Schematic representation of MRI tracking of SPIO labeled cells. Cells are labeled with SPIO nanoparticles via co-incubation, expanded and then injected intravenously. The particles can then be detected via MRI imaging.¹²¹ Copyright © 2022 The Authors. Advanced Functional Materials published by John Wiley and Sons. (B) MRI images of SPIO labeled cytotoxic cells accumulated at the tumor site and in lymph nodes in response to the combination of the DPX immunotherapy vaccine and anti-PD-1 immunotherapy.¹²³ © 2020 The Author (s). Published with license by Taylor & Francis Group, LLC. (C) Representative MRI images of mice treated with ¹⁹F labeled splenocytes or T cells. 19F-MRI is represented in orange, which 1H-MRI is in grey.⁸⁰ © 2016 Gonzales et al.

Non-radioactive fluorine (¹⁹F) can also be used as a contrast agent for MRI. ¹⁹F has benefits over other contrast agents in that it produces high contrast in labeled cells due to the low background ¹⁹F signal in other tissues.⁷⁸ Furthermore, it has been demonstrated to be non-toxic, requires only small doses to achieve sufficient signal, and does not experience radioactive decay, making it ideal for in vivo longitudinal monitoring.^{79,125 19}F has been used to track therapeutic dendritic cells from various sources, including bone marrow and fetal skin. In both cases, cells were efficiently labeled with a ¹⁹F perfluoropolyether (PFPE) nanoemulsion ex vivo and then injected into mice, either directly into tissues or intravenously, and tracking was successful with minimal impact on DC functions.⁷⁸ A similar labeling technique was used on NK cells used in adoptive cell therapy. NK cell viability and cytotoxicity was maintained after labeling with a ¹⁹F PFPE nanoemulsion. Labeled NK cells were seen migrating up to 8 days following an injection, demonstrating the potential of ¹⁹F for longitudinal immune cell tracking.⁷⁹ Emulsions containing ¹⁹F-based perfluorocarbons have also been used to label T cells and splenocytes to monitor migration in a melanoma model, and signal was detected in the liver, lungs and spleen. Signal was higher in splenocytes as compared to T cells due to comparatively higher proliferation of T cells, which diluted the signal in daughter cells (Fig. 5c).⁸⁰ This represents a challenge to the use of ¹⁹F as a contrast agent for immune cell tracking, and to most contrasts developed using the materials engineering approach.

Chemical exchange saturation transfer (CEST) MRI is an MRI technique that enables the detection of compounds at concentrations lower than would be detectable by standard MRI, and it represents a potential method of addressing diluted signals due to cell proliferation. In CEST, chemical species are able to exchange ¹H protons with water. A resonant radiofrequency is applied that causes the chemical species to reach a saturation state, and over time, this saturation is transferred to water as protons are exchanged. This results in decreased water signal, which enables increased signal of the chemical species in comparison.¹²⁶ While endogenous contrasts for CEST MRI can be used, exogenous contrasts can be predictably controlled, improve sensitivity and specificity, and increase signal-to-noise ratio compared to endogenous contrasts. Exogenous contrasts have been designed based on paramagnetic materials, liposomes and iodine, though more research must be done to establish their efficacy for immune cell tracking.¹⁰²

3.5. Photoacoustic imaging

Photoacoustic (PA) imaging is a relatively new modality of considerable interest, due to its non-invasive, non-ionizing nature, fine spatial resolution, and capacity for multiplexed monitoring of multiple biological processes. It is also easy to integrate into clinics and has benefits over other imaging modalities, as it is portable, cost-effective, efficient, and familiar to staff. Furthermore, it is frequently used in combination with ultrasound, which provides deep tissue anatomical information.^82,127,128 PA imaging has also shown potential in therapeutic interventions, including light-based therapies, such as photodynamic therapy (PDT) and photothermal therapy (PTT).^86,127

PA imaging uses the photoacoustic effect to generate signals. In the photoacoustic effect, an acoustic wave is generated using a pulsed laser. The pulsed laser excites electrons, which release heat as they relax back to the ground state. The released heat causes thermal expansion and generates an acoustic wave. PA imaging incorporates a laser to generate this signal along with an ultrasound probe to detect the resulting acoustic wave (Fig. 6a). While PA signal can be generated with endogenous contrasts, such as hemoglobin and melanin, it can be further enhanced via exogenous contrasts. Furthermore, the use of exogenous contrasts enables the visualization of cells and biological processes that would not otherwise be seen with PA imaging (Fig. 6b and c).¹²⁹

Fig. 6.

PA imaging for in vivo immune cell tracking. (A) Schematic depiction of PA imaging system. (B) Absorption spectra of endogenous contrasts in the body. (C) Absorption spectra of common exogenous contrasts¹²⁸ Copyright © 2021 The Authors. Trends in Biotechnology published by Elsevier. (D) In vitro PA images of tissue mimicking dome phantoms with varying concentrations of gold nanorod labeled cells. Higher concentration of nanorod-labeled cells corresponds with a higher PA signal intensity.⁸³ © 2013 Optical Society of America (E) In vivo PA images of gold nanorods (AuNR@SiO₂) and nanorod labeled CIK cells (NP-CIK) delivered via peritumoral injection. Signal was detected up to day 4, demonstrating the use of PA imaging for longitudinal noninvasive, in vivo tracking of immune cells.⁸⁵ © Yang et al. 2016.

Exogenous contrasts for PA imaging can be classified into several groups, including metal-based nanoparticles, carbon-based nanoparticles, small molecule dyes, and semiconducting nanoparticles.^130,131 For example, Zheng et al. demonstrated the use of the small molecule dye NIR-797-isothiocyanate to label T cells for visualization with PA imaging. T cells labeled with NIR-797-isothiocyanate were adoptively transferred to mice, and their accumulation at the tumor site was able to be detected non-invasively over time.¹³⁸ Semiconducting polymer nanoparticles have also been used for immune cell tracking, though not specifically in the context of immunotherapy. In one study, macrophages, neutrophils and MSCs were labeled with semiconducting polymer nanoparticles composed of thiadiazoloquinoxaline, enabling the monitoring of cell behavior in a model of inflammation.⁸¹ Given the successful labeling of cell types commonly used in immunotherapy, this approach could be readily adapted for tracking immune cell targeting of tumors.

Gold-based nanoparticles are among the most commonly used contrast agents for PA imaging in cell tracking applications, as they do not interfere with immune cell biological activity and provide a strong PA signal. Additionally, they can be made in various shapes to modify optical properties, and with various surface modifications to optimize imaging of specific tissues or biological processes.⁸³ In one study, gold nanorods were used to label dendritic cells that were stimulated with tumor cell-derived exosomes. Piao et al. investigated tumor cell-derived exosomes as a method of encouraging DC activation and maturation for immunotherapy. One marker of DC activation is migration to the lymph nodes, which was successfully monitored using PA imaging. This study demonstrates the potential of PA imaging for aiding in the development of immunotherapies.⁸²

PA imaging is also commonly combined with fluorescence imaging due to the high sensitivity and specificity fluorescence imaging provides.¹³² These benefits, in combination with the increased tissue depth penetration offered by PA imaging relative to fluorescence imaging, enable the tracking of multiple biological processes and cell behaviors using one contrast agent. Gold nanoparticles are a popular contrast agent for bimodal imaging as well. In one study, gold nanorods were coated in fluorescent dye doped silica and used to label macrophages. Silica coating was determined to increase stability and signal strength, while fluorescent dye enabled easy visualization of labeled cells and facilitated ex vivo imaging. PA imaging of the nanorod labeled macrophages demonstrated high sensitivity, providing a strong basis for the use of bimodal imaging for tracking of immune cells (Fig. 6d).⁸³ A more recent study successfully demonstrated the application of silica-coated gold nanorods for tracking T cells in the context of adoptive T cell transfer. This approach enabled the in vivo monitoring of T cell trafficking using ultrasound-guided PA (US/PA) imaging, further highlighting the potential of combining nanoparticles with PA imaging for immune cell tracking and therapeutic monitoring.⁸⁴

Contrasts designed for use with PA imaging can also act as a therapeutic via the use of PTT. PTT involves the heating of photothermal agents via visible or non-visible wavelengths, leading to the destruction of cancer cells due to their increased heat sensitivity.¹³³ Gold nanoparticles can be used as photothermal agents. Silica coated gold nanorods were used to label cytokine-induced killer (CIK) cells, which are a type of T cell. CIK cells are used as an immunotherapy for treatment of gastric cancer, and labeling with gold nanorods increased their efficacy by triggering the release of cytokines that induce apoptosis in gastric cancer cells. Labeling with gold nanorods also enabled cell visualization using PA imaging, as well as further treatment via PTT (Fig. 6e).⁸⁵ Similarly, CIK cells were labeled with PEGylated gold nanoprisms for photoacoustic imaging, enhanced immunotherapy and photothermal therapy. Optimal synergistic effect was seen using a combination of immunotherapy for the first three days, followed by photothermal therapy.⁸⁶

These applications of photoacoustic imaging demonstrate the versatility of the modality, as it can be used in conjunction with other imaging modalities with ease, while also providing therapeutic benefits via PTT. Furthermore, multiplexed PA imaging offers the ability to measure various parameters at the same time by taking advantage of the ability of PA imaging to track multiple endogenous and exogenous contrasts at once (Fig. 6b and c).^128,134 If the contrasts of interest have different absorption spectra, they can be visualized simultaneously after excitation with a range of wavelengths of light.¹³⁴ This enables tracking of parameters such as oxygenated and deoxygenated hemoglobin, as well as tumor cells, immune cells and more, all of which can provide information about tumors and the progress of immunotherapy treatments.^128,135 The ability to image multiple parameters at once results in less experimental variability when developing contrast agents.¹³⁴ Furthermore, the ability to track multiple parameters at once with one diagnostic could save time and be more cost effective when it comes to diagnosis and treatment of cancers in clinical settings.

One issue with PA imaging that must be addressed in order to achieve successful clinical translation is the limited penetration depth. In order for PA signal to be generated, light must travel through the tissue to excite electrons.^129,136 However, light from the pulsed laser attenuates quickly in these tissues, which limits the depth of PA imaging. A potential solution for this issue is the development of optical fibers for minimally invasive PA imaging. Optical fibers are integrated in a medical device such as a needle or endoscope, which allows light to be delivered directly to the target. Optical fibers integrated with needles can be inserted percutaneously and signal can be detected by an external ultrasound probe, which is known as interventional PA imaging. PA endoscopy, another option for deep tissue imaging, uses a probe with a combined optical fiber and ultrasound detector, which can either be inserted percutaneously or into hollow structures such as veins and arteries.¹³⁶ While PA imaging does face challenges in clinical translation, further development of imaging systems and contrasts agents give it the potential to circumvent these issues and act as a powerful diagnostic and therapeutic tool for treatment of cancers.

4. Summary and prospects

Cell-based immunotherapies have become a popular method for cancer treatment due to the intrinsic ability of immune cells to target cancer cells and the improved outcomes as compared to other treatment methods.^2,4 Despite the successes seen with immunotherapy, several challenges persist, including limited understanding of its mechanisms, difficulty in predicting treatment efficacy, and the possibility for adverse side effects.^17–19 These limitations necessitate a non-invasive, real-time, and longitudinal monitoring approach for immune cells used in immunotherapy.

Conventional monitoring approaches include tumor tissue biopsies that are invasive and do not provide accurate information on the infiltration of the treatment or the distribution of the cell based therapy on a whole body level.¹¹⁴ Imaging methods such as optical imaging, CT, MRI, PET/SPECT and PA imaging have the ability to visualize in vivo processes with great accuracy compared to traditional monitoring approaches. These imaging techniques have provided novel mechanisms by which to understand the mechanism and track the efficacy of cell-based therapies.

Nonetheless, improvements to existing non-invasive imaging approaches remain necessary for each modality. While current techniques typically evaluate therapeutic outcomes in the short term, long-term in vivo tracking is essential to accurately assess treatment efficacy. Achieving this requires the development of durable tracers that persist without exerting toxic effects on immune cells. Additionally, the integration of interdisciplinary techniques, such as microfluidic platforms from mechanical engineering, has the potential to enhance the efficiency of cell labeling while minimizing tracer-induced cytotoxicity.¹³⁷

Moreover, PA imaging presents a unique opportunity for multiplexed imaging, distinguishing it from other in vivo imaging modalities. Unlike CT, MRI, or PET/SPECT, PA imaging can simultaneously capture both structural and functional data, such as oxygen saturation within the tumor microenvironment, while tracking immune cell distribution as demonstrated in a recent study.⁸⁴ This dual capability allows researchers and clinicians to directly correlate immune cell behavior with dynamic changes in tumor physiology, enabling real-time assessments of treatment efficacy. Moving forward, standardizing PA imaging protocols and integrating the technology into preclinical and clinical workflows will be critical to fully leveraging its potential for optimizing cell-based immunotherapies, and we are actively working on these advancements.

However, despite its advantages, PA imaging remains limited by the relatively low tissue penetration depth of light during transillumination from the skin. Solutions to mitigate this involve the insertion of optical fibers into the body to deliver light directly to the target; however, while these approaches are minimally invasive, they remain inherently invasive procedures.¹³⁶ These constraints highlight the need for continued research into better non-invasive, in vivo tracking modalities. Advancements in this field will facilitate long-term, non-invasive imaging to guide therapy optimization and offer patient-specific insights into treatment outcomes.

Biographies

Anika Kulkarni is a Ph.D. student in Biomedical Engineering at the University of California, Davis, and a Graduate Student Researcher in the Laboratory for Cell Nanoengineering. She earned her Bachelor of Science degree in Bioengineering from the University of California, Los Angeles in 2023. Her research focuses on the development of functional nanomaterials for theranostic applications in bone regeneration and musculoskeletal diseases. Specifically, she is designing functional nanoparticles and contrast agents for noninvasive monitoring and modulation of cell behavior in cell therapies using ultrasound and photoacoustic imaging.

Tasneem Mukarrama is a post-baccalaureate researcher and an incoming PhD student in Immunology in the Laboratory for Cell Nanoengineering at the University of California, Davis. She earned her bachelor’s degree in Biochemistry and Molecular Biology from the University of California, Davis in 2023, where she developed an interest in immunotherapy research. Her work focuses on advancing NK cell-based therapies for solid tumor malignancies, with an emphasis on improving treatment efficacy and persistence. She also explores non-invasive imaging techniques, such as ultrasound and photoacoustic imaging, to enhance the tracking and monitoring of these therapies in vivo.

Brendan Barlow is a post-baccalaureate researcher and an incoming PhD student in Integrative Pathobiology in the Laboratory for Cell Nanoengineering at the University of California, Davis. He graduated with a Bachelor of Science degree in Biochemistry and Molecular Biology from the University of California, Davis in 2023. His research focuses on the non-invasive tracking of immune cells in vivo through the development and application of nanomaterials for cancer therapeutics. His work spans a range of applications, from small animal models to companion animals such as dogs, aiming to advance translational approaches for immune cell-based therapies.

Dr Jinhwan Kim is an Assistant Professor in the Department of Biomedical Engineering and the Department of Surgery at the University of California, Davis, where he serves as the Principal Investigator of the Laboratory for Cell Nanoengineering. He earned his Ph.D. in chemistry from POSTECH and completed his postdoctoral training in bioimaging at Georgia Tech. His current research focuses on the design and application of functional nanomaterials for theranostic applications, with an emphasis on engineering therapeutic cells with nanomaterials and integrating non-invasive imaging techniques to enhance cell-based therapies. His work spans a range of biomedical applications, including regenerative medicine and cancer therapy, with the goal of advancing translational strategies for precision medicine.

Data availability

This manuscript does not include the generation of any new data.