From Theranostics to Immunotheranostics: the Concept

Abstract

Immunotheranostics will be an important development in the future of nuclear medicine and medical oncology. It describes the synergy of theranostic procedures in nuclear medicine and immune oncology (IO) treatment. In brief, it takes advantage of molecular imaging and subsequent targeted modulation of the—in most cases immunosuppressive—tumor microenvironment (TME) by diagnostic and therapeutic radioisotopes. This is of high importance since only a fraction of patients receiving IO is currently being cured by this exciting therapy option. We therefore envision the concept of immunotheranostics as a powerful mean to augment the success of IO treatment in the future and thus the urgent need to further develop the interaction and joint action of nuclear medicine and medical oncology for substantially improved therapy outcome for cancer patients.

Introduction

Nuclear medicine and especially theranostic procedures have gained attention in the last years and appear to be on the fast lane of success. The theranostic principle—to visualize a disease target by nuclear medicine/molecular imaging methods in its whole-body expression prior to a systemic treatment directed at the identical target—appears to be evident and promising. With the tremendous success of the NETTER-1 trial [1] in well-differentiated midgut neuroendocrine tumors and the impressive clinical results of lutetium-177- and actinium-225-based prostate-specific membrane antigen (PSMA) treatments, there is rising clinical attention and exposure of these promising therapeutic methods [2, 3].

These important successes also leveraged nuclear medicine as one of the most promising partners in the oncological treatments of cancer patients. However, in parallel to the evolution in targeted radionuclide therapy, intense research in immune oncology (IO) brought forth a variety of different checkpoint inhibitor treatments or other immune-modulating anti-cancer agents. It is an open question, if and how these two worlds, which seem to develop in parallel as very promising methods, can be used synergistically and learn from each other.

Recently, several high-ranking studies and trials using nuclear imaging methods to non-invasively visualize and quantify the targets of IO treatments have been published by several collaborative groups. A first-in-human study of de Vries et al. showed a generally high, but heterogeneous uptake of zirconium-89-labeled atezolizumab (anti-PD-L1) in tumor patients [4]. Very importantly, the clinical responses were correlated with the pretreatment PET signal more than with immunohistochemistry or RNA-based sequencing. A different group used an F-18-labeled anti-PD-L1 Adnectin (F-18-BMS-98619) which showed a very favorable tumor imaging and favorable in-human distribution [5]. The same study showed a significant prediction of patients responding to checkpoint inhibitor treatment versus non-responders using the labeled Adnectin and the Zr-89-based immuno-imaging. The results of both first-in-human clinical studies are highly interesting and show the great potential of direct imaging of the target of checkpoint inhibitor treatment.

Despite these very interesting feasibility studies, a provocative question remains: Whether these imaging markers will progress into clinical routine. First, and most importantly, the biological background of response to checkpoint inhibitors is very complex. There are many different direct and indirect barriers to an effective immune response. Expression of the PD1/PD-L1 target does therefore not necessarily lead to a meaningful response, especially if checkpoint inhibitor treatment is used over a longer period of time, leading to T cell exhaustion. This has been explained by upregulation of inhibitory T cell receptors such as TIM-3 or LAG-3 and many others, which hampers an effective immune response, even in presence of PD-1 expression [6], but also by the action of other antagonistic factors in the so-called tumor microenvironment. The multifaceted interplay between agonistic and antagonistic effects makes therefore a classical theranostic “find the target–treat the target” paradigm relatively complex and maybe not as successful as thought.

PD-L1/PD-1 PET Imaging

Whole-body imaging of the target receptor, as for example PD-L1/PD-1, might be an interesting step in the right direction. However, the clinical introduction of such a procedure remains complex on several levels. First, patients would need repeated imaging to understand evolving resistance, and recurrent Zr-89-based imaging is anticipated to add an important radiation burden to the patient (~ 20 mSv per administration) [7]. Secondly, the price of an Zr-89-based PET scan or proprietary diagnostic radiopharmaceuticals (e.g., F-18-BMS-98619) will be considerable, and thus the economic value, i.e., to answer the question whether we can exclude a patient from IO according to the result of the PET scan, needs to be answered in prospective clinical trials with a relatively complex design. Third, this would require establishing a strong relationship between medical oncology and nuclear medicine, including cross training in both fields to facilitate the information flow and the understanding of relevant questions by the respective other group [8, 9]. Further questions touching reproducibility of imaging results, including phantom measurements for each PET scanner, as well as standardized reporting, also need to be addressed in prospective trials, ideally within large collaborative groups.

Overall, we believe these are very important questions, which can only be answered successfully if either we highly augment the clinical effectiveness or drastically reduce overall cost of IO treatments by excluding the right patients based on valid and reliable molecular biomarker imaging results.

From Theranostics to Immunotheranostics

To address the above problem, we have to think back of what made theranostics successful. It is not the diagnostics; it is the therapy in “THERA”nostics, which made it a strong principle. So it is not sheer coincidence that in the art term theranostics, “theran” appears before “ostics”.

We strongly believe that immunotherapy is the most relevant and powerful modern treatment in oncology, since it can induce true cancer cure in selected patients. Thus, it seems a next logical step to add “Immuno” before “Theranostics” and introduce the art term “Immunotheranostics” (Fig. 1). For us, this implies not only the imaging validation of the target as pointed out above—meaning to find an IO target—but also to modulate an IO target to make it more accessible for immunotherapy, which ultimately will lead to a higher percentage of cure of patients by IO (Fig. 2). We strongly believe that immunotheranostics will play an important role to overcome the riddle of ineffective IO therapy in the context of the antagonistic TME and its plasticity and there is room for true synergism.

Fig. 1.

Immunotheranostics combines synergistically the principle of immunooncology and theranostics by imaging and modulating the tumor microenvironment. This Sankey diagram also reflects the relative proportion of articles published in PubMed in each field in 2019 (dark blue and orange; left)

Fig. 2.

Partly hypothetical Kaplan-Meier curves showing the possible impact of personalized immunotherapy which is guided by patient selection, e.g., by multiple biopsy results or whole-body PET immune-imaging and of immunotheranostics which aims to visualize and therapeutically target the immunosuppressive tumor microenvironment

Immunotheranostics is of course much more complex than classical theranostics and needs more in-depth biological understanding of a changing heterogeneous cellular, local, and possibly systemic target over time. However, especially in this challenge, we see one of the main opportunities in nuclear medicine providing whole-body IO molecular imaging and therapeutic systemic radiopharmaceuticals for the following years.

Immunotheranostics: Imaging and Modulating the Tumor Microenvironment

Of course, T cells are the central players in immunooncology, their presence and activation state in the tumor being pivotal for success or failure of immune checkpoint inhibition. Several PET imaging agents allowing the visualization and activity of specific T cell populations (CD8⁺, CD4⁺) [10] are currently under clinical and preclinical investigation and indeed hold the promise for an improved and more specific assessment of therapy response towards cure. However, imaging of the “master players” in antitumor-immunity only provides information on a significant, but singular aspect of the highly complex tumor biology, and taking into account the entire context of the tumor microenvironment with its diverse and highly dynamic immune infiltrate is a prerequisite for understanding the functioning of an immune suppressive TME and thus, ultimately, for efficient modulation of antitumor response by immunotheranostics.

Tumor-Associated Macrophages

For example, macrophages are present in a fragile equilibrium of pro-inflammatory and anti-inflammatory subtypes, including the so-called tumor-associated macrophages (TAMs), which usually are suppressing an effective cellular immune response. TAMs promote tumor progression and are mainly present in advanced, invasive stages of cancer development. One of the relevant TAM targets is the mannose receptor (“Cluster of Differentiation” 206 or CD206). Besides expression on TAMs, CD206 is mainly expressed in immature dendritic cells and liver sinusoidal endothelial cells. It has been shown that expression of CD206 on macrophages drives them to produce IL-10 and TGF-beta, two very relevant inhibitors of an active cellular immune response. Liu et al. used CD206 antibody SPECT imaging for early prediction of post-chemotherapy tumor relapse in a preclinical breast cancer model [11]. Another group used a CD206 binding, F-18 bound single-domain camelid to detect CD206-positive macrophages in tumoral lesions [12]. To the best of our knowledge, there is no open human trial investigating the role of CD206 imaging in cancer patients. However, CD206 might be an interesting target in the future to predict outcome of immunotherapy, as well as to predict and visualize early resistance to therapy. A useful application of CD206 might be an immunotheranostic approach, however. Macrophages are relatively radiosensitive, and it has been shown that already very low-dose radiation can reprogram immunosuppressive macrophages into a pro-inflammatory phenotype and therefore augment the efficacy of IO treatment [13].

Cancer-Associated Fibroblasts

Furthermore, the highly selective targeted imaging of tumoral fibroblasts (cancer-associated fibroblasts or CAFs), using selectively upregulated fibroblast activation protein (FAP) as a molecular target, has recently gained enormous attention. Specifically, a series of radiolabeled FAP inhibitors (FAPI) has been extensively used and proposed as a new and quite universal diagnostic method for cancer imaging [14]. However, it needs to be taken into account that the FAPI imaging signal might probably not reflect the spread of cancer, but rather indicates the distribution of the FAP-positive CAFs, which is a sign of advanced desmoplastic tumors as a response to the challenging microenvironment. FAP, a constitutively active serine peptidase, promotes malignant and invasive behavior with malignant cells using a complex cross talk of cancer cells and local fibroblasts, and its pharmacological targeting leads to suppression of pro-tumoral activity [15]. Despite all this important biological importance of FAP, the distribution of CAF might not reflect entirely the actual tumor spread, but evolving CAF-mediated resistance including possible chronic inflammatory processes, e.g., chronic pancreatitis. It might be therefore less relevant for imaging and staging unless we have a specific clinical question. However, the use of FAP in a theranostic setting might be even more interesting [16]. However, CAFs are a very relevant barrier to IO therapy by inhibiting the cellular immune response by secretion of many antagonistic cytokines (e.g., CXCL12, CXCL8, IL-6, tumor necrosis factor, TGF-beta, and VEGF), extracellular matrix remodeling, or suppression of NK cell activity [17]. Disrupting the FAP-positive CAF barrier might be of the highest interest in synergy with IO treatment.

Tumor Neoangiogenesis

Prostate-specific membrane antigen (PSMA) is probably the most exploited and most cited theranostic target of the last years. Clinical case studies as well as phase II theranostic trials have shown impressive results [18, 19]. Besides in prostate cancer, PSMA also plays an important role in cancer neoangiogenesis. Studies have shown a poor prognosis, e.g., in patients with HCC, where PSMA overexpression was observed in the tumor lesions [20]. Since PSMA facilitates endothelial invasion via β1-integrin signaling and actin-binding protein Filamin A, this observation may be linked to increased tumor angiogenesis in these lesions [21]. Importantly, tumor neoangiogenesis is one of the major drivers of resistance to immunotherapy. Activated endothelial cells can reduce T cell activity, tag them for destruction, and block them from gaining entry into the tumor for example by expression of antagonistic receptors or secretion of antagonistic cytokines [22]. Targeting PSMA-positive tumor vasculature using for example lutetium-177-labeled PSMA ligands as a vascular disruption agent might therefore, especially in combination with immunotherapy, be a very interesting option in the future.

These three examples show that immunotheranostics is not aiming at nuclear targeting of surface molecules overexpressed on cancer cells but tries to modulate an IO-associated target to render the microenvironment more sensitive for IO treatment, either by destructing a certain cell type, which functions as immune barrier or by generally “inflaming” a tumor, or to augment local and distant effects synergizing with immunotherapy [23].

Conclusion

Overall, we believe that immunotheranostics will be the major advancement in nuclear medicine–driven oncology. It takes advantage of molecular imaging and subsequent targeted modulation of immunosuppressive barriers to IO by diagnostic and therapeutic radioisotopes to augment the efficacy of checkpoint inhibitor treatments. Due to the intra-individual heterogeneity and over time variability of these IO barriers, it will open a new paradigm on how we will treat cancer in the next years in a truly synergistic way by (A) a biological mechanistic synergism of target identification and target modulation by combinations of radionuclide therapy and immunotherapy, (B) hopefully, an increasingly efficient collaboration of nuclear medicine specialists with medical oncologists, and (C) involving and exciting our industrial partners for such large-scale multidisciplinary programs.

Compliance with Ethical Standards

Conflict of Interest

Niklaus Schaefer, John O. Prior, and Margret Schottelius declare that they have no conflict of interest.

Ethical Approval

This article does not contain any studies with human participants performed by any of the authors.

Informed Consent

Not applicable.