Bench-to-bedside translation of interleukin-15 for immunotherapy: principles and challenges

1. Introduction

Since its discovery 25 years ago by our group and Grabstein et al. [1–3], interleukin-15 (IL-15) has attracted much attention in terms of its biological activities, mechanism of action, and therapeutic modulations under various physiological and pathological conditions. In particular, the stimulatory effects of IL-15 on the proliferation and effector functions of natural killer (NK) cells and CD8⁺ T cells make it a promising component for immunotherapy in the treatment of malignancies and infectious diseases [4,5]. Here, we present a brief summary on the biological principles of how IL-15 exerts immunostimulatory effects on cells, the clinical investigations on IL-15-based therapeutic agents, and the challenges for safe and efficacious delivery of IL-15 in patients.

2. IL-15, a critical cytokine for NK and CD8⁺ T cell expansion

IL-15 is a pleiotropic cytokine that stimulates the differentiation and proliferation of NK cells and T cells [6–11]. IL-15 signals through a heterotrimeric receptor including the common gamma chain (γ_c) shared with IL-2, IL-4, IL-7, IL-9, and IL-21, the β chain shared with IL-2, and a unique IL-15Rα chain. IL-15 and IL-15Rα are produced in a coordinated fashion mostly by activated monocytes, macrophages, and dendritic cells. Stimulation of these cells with type I/II interferons, or NF-κB activation through CD40 ligands (CD40 L) or Toll-like receptor (TLR) agonists, induces the synchronized expression of IL-15 and IL-15Rα. An unusual aspect of IL-15 and IL-15Rα interaction is the extremely high binding affinity (~5 × 10⁻¹¹ M). IL-15 is only secreted in a small quantity and is mainly membrane-bound under physiological conditions. IL-15 and IL-15Rα expressed by monocytes and dendritic cells are associated on the cell surface and can recycle through endosomal vesicles for many days, resulting in persistence of membrane-bound IL-15Rα and the associated IL-15. IL-15 bound to IL-15Rα presents IL-15 in trans to cells express IL-2/15Rβγ_c but minimal IL-15Rα [12]. Such IL-2/15Rβγ_c-expressing target cells by the IL-15–IL-15Rα pair include NK cells and CD8⁺CD44^hi memory T cells.

IL-15 is one of the most important cytokines that regulates the generation, persistence and differentiation of NK cells and CD8⁺ T cells. Although sharing two receptor components (IL-2/15Rβγ_c), IL-2 and IL-15 make distinct contributions to adaptive immune responses. As the first FDA-approved immunotherapeutic agent, IL-2 supports among others the cytotoxic capacity of CD8⁺ T cells and promotes naïve CD4⁺ T cell differentiation. IL-2 through its induction of regulatory T cells (Tregs) and activation-induced cell death (AICD) acts as an immunological checkpoint required for self-tolerance [13,14]. Moreover, IL-2 is known to cause capillary leak syndrome and other severe toxicities, which can negate the therapeutic benefits, as revealed in a recent clinical trial that involved IL-2 in a combination immunotherapy [15]. In contrast, IL-15 inhibits AICD and favors the survival of CD8⁺ memory T cells, and is thus dedicated to supporting the persistence of immune responses [4,16]. On the basis of these distinct characteristics, it has been postulated that IL-15 may be superior to IL-2 in the treatment of malignancies and as a component of vaccines against viral infections such as influenza and HIV.

3. Clinical trials with IL-15

IL-15 has emerged as one of the highly promising candidates in immunotherapy and is currently undergoing clinical trials in patients with malignancies. In therapeutic settings, IL-15 alone can bind to the intermediate-affinity IL-2/IL-15Rβγ_c receptor complex in the absence of the high-affinity IL-15Rα, resulting in the activation of downstream signaling in lymphocytes. Clinical grade Escherichia coli-produced recombinant human IL-15 (rhIL-15) has been manufactured under the GMP conditions by the Biopharmaceutical Development Program at NIH [17]. Three clinical trials of rhIL-15 (NCT01021059, NCT01727076, NCT01572493) have been completed in patients with advanced cancers [18–20]. In light of strong additivity/synergy observed with IL-15 in combination with monoclonal antibodies including immune checkpoint inhibitors, more clinical trials of rhIL-15 in combination therapies have recently been initiated (NCT02689453, NCT03388632, NCT03759184, NCT03905135, NCT04150562, NCT04185220).

In the first-in-human phase I clinical trial of rhIL-15, treatment was given as an intravenous bolus (30-min IV infusion) once daily for 12 consecutive days [18]. The maximum-tolerated dose (MTD) was found to be 0.3 μg/kg/day, and further dose escalation was constrained by post-infusion toxicities. Subsequently, a second phase I clinical study was conducted by subcutaneously (SC) administering rhIL-15 once daily for 10 days, and the MTD was determined to be 2 μg/kg/day [19]. The SC bioavailability of rhIL-15 was close to 20%, indicating reasonable stability and absorption of rhIL-15 in the subcutaneous milieu. In the third trial, rhIL-15 was administered via continuous IV infusion for 10 days, and the MTD was found to be 2 μg/kg/day [20]. Both SC and continuous IV infusion treatment of rhIL-15 were usually well tolerated in patients with mild side effects. Pharmacokinetic (PK) analysis showed that rhIL-15 was rapidly eliminated from the systemic circulation following IV injection, whereas the serum level of rhIL-15 following SC injection was more protracted owing to the prolonged absorption phase of rhIL-15; and the serum concentration was sustained over 10-day continuous IV infusion (Table 1). As pharmacodynamic (PD) markers for in vivo activity of IL-15, the circulating NK cells and CD8⁺ T cells were monitored in all trials. Continuous IV infusion of rhIL-15 resulted in the most drastic increase in the circulating effector cells − 38-fold increase in NK cells with 358-fold increase in CD56^bright NK cells and sixfold increase in CD8⁺ T cells, while IV bolus administration being the least effective (Table 1). Counteracting of transforming growth factor β1 (TGF-β1) signaling by IL-15, a potent immunosuppressor in NK cells, may play a role in the drastic NK activation [21,22]. It is evident that the magnitude of immune responses to IL-15 was correlated with the persistent exposure of IL-15 rather than the peak serum level, underscoring the importance of sustaining the serum IL-15 level.

Table 1. PK/PD of IL-15-based therapeutic agents in phase I clinical trials.

	MTD or highest dose tested	Cmax (ng/ml)	Duration for IL-15 serum level ≥ 0.1 ng/ml	Maximum fold increase in total NK cells	Maximum fold increase in CD56^bright NK cells	Maximum fold increase in total CD8^+ T cells	Ref
rhIL-15	0.3 μg/kg/d IV bolus for 12 days	1.3	5 h	2–3	3–4	3	[17]
	2 μg/kg/d SC on days 1–5 and 8–12	1.6	10 days	10.8	39.7	3.3	[18]
	2 μg/kg/d continuous IV infusion for 10 days	2.9	10 days	38	358	5.8	[19]
ALT-803	10 μg/kg IV or SC once-weekly	IV: 200 SC: 0.5	1 day per week 4 days per week	3 8	2 8	2 2	[20]
	6 μg/kg IV or 20 μg/kg SC once-weekly	IV: 90 SC: <1		2.5 3.3	5.3 6.3	1.4 3.1	[21]

ALT-803, an IL-15 ‘superagonist’ complex consisting of IL-15 mutein (N72D) bound to the ‘sushi’ domain of IL-15Rα fused to human immunoglobulin IgG1 Fc, is another IL-15 therapeutic that is currently undergoing clinical trials in cancer patients. In the first-in-human phase I trial [23], ALT-803 was IV and SC administered to patients once-weekly at doses up to 10 μg/kg (molar equivalent to 2.8 μg/kg IL-15). IV administration of ALT-803 caused post-infusion toxicities (fever and chills/rigors), which coincided with notably increased serum levels of IL-6 and IFNγ. Such adverse effects were largely mitigated by changing the dosing route to SC administration. However, SC dosing of ALT-803 caused large injection site rashes (erythematous plaques) owing to the infiltration of CD56⁺NKp46⁻ γδ T cells. In contrast to the high peak serum level and gradual decline within 24 h following IV dosing, the serum ALT-803 level peaked several hours after SC injection with the C_max 100-fold lower than that of the IV route, which was maintained at the steady state for 96 h. SC administration of ALT-803 resulted in eightfold increase in NK cells and twofold increase in CD8⁺ T cells (Table 1). Similar PK/PD findings were observed in a second phase I trial in patients with advanced solid tumors [24]. In another recent phase I trial, combination therapy of ALT-803 and nivolumab was found to be well tolerated in non-small cell lung cancer patients [25].

4. Challenges for sustained delivery of rhIL-15

PK/PD results from the above clinical trials highlight the importance of maintaining steady serum IL-15 level for weeks in order to stimulate prolonged expansion of NK cells and CD8⁺ T cells. Nevertheless, expansion in the effector cells induced by ALT-803 is notably less pronounced compared to rhIL-15. The development of large erythematous plaques following SC injection of ALT-803 in patients makes further dose escalation difficult. On the other hand, continuous IV infusion of rhIL-15 is problematic in terms of patient compliance, quality of life, and medical cost. It is thus imperative to improve the dosing strategy to optimize the outcomes of IL-15 therapy.

As a recombinant protein with molecular size of 14–15 KDa, rhIL-15 undergoes rapid elimination with a brief half-life of 2.7 h in humans [18]. Continuous IV infusion renders a constant rate of drug input into the systemic circulation that offsets drug elimination, resulting in a persistent drug level at the steady state. From the standpoint of drug delivery, it is highly desirable to develop a controlled-release formulation for rhIL-15 that mimics continuous drug input with zero-order kinetics offered by IV infusion while minimizing the dosing frequency. We envision that in situ forming injectable hydrogels may be tailored as such a delivery vehicle for rhIL-15 that would release the protein in a controlled manner. Such hydrogels may be injected noninvasively through a standard syringe in aqueous solution form that is subsequently solidified in vivo. Hydrogels are an appealing matrix for protein delivery owing to their large aqueous content, allowing for facile protein loading under mild conditions without denaturation or loss of bioactivity [26]. The spatiotemporal control over drug delivery could be achieved by tuning the physicochemical properties of the hydrogel components. These materials could be derived from both natural and synthetic sources, which need to be biocompatible and biodegradable for clinical applications. We are currently exploring once-weekly injection of an rhIL-15 depot formulation that employs poly(lactide-co-glycolide)-poly(ethylene glycol)-poly(lactide-co-glycolide) (PLGA-PEG-PLGA)-based thermosensitive hydrogel as a delivery system. Compared to the standard drug solution, a single SC injection of rhIL-15 hydrogel was found to prolong the half-life of rhIL-15 by at least 10-fold in murine models, resulting in an elevated drug level for days without repeated administration. The ultimate goal of such sustained-release rhIL-15 delivery system is to replace continuous IV infusions and to achieve an equivalent magnitude of expansion in NK cells and CD8⁺ T cells in vivo. Alternatively, gene delivery approach has been experimented to achieve elevated IL-15 level in the circulation, and the proof of concept for such an approach was demonstrated by intramuscular injection of DNA vectors encoding heterodimeric IL-15 in rhesus macaques [27].

5. Expert opinion

Defining the optimal treatment regimens is essential for developing cytokines as cancer immunotherapeutics. Given its pivotal role in the expansion and effector functions of NK cells and CD8⁺ T cells, IL-15 has been intensely pursued as an immunotherapeutic agent to augment innate and adaptive immune responses against neoplasia and pathogens. Successful clinical translation of IL-15 will hinge upon the optimization of therapeutic regimens and delivery strategies, that allow for safe and sustained IL-15 levels without excessive release of inflammatory cytokines. The development of a controlled release long-acting IL-15 depot that mimics continuous drug infusion would be an important goal. Nevertheless, all forms of IL-15 despite their dramatic augmentation of the number and activation status of NK cells and CD8⁺ T cells when administered as monotherapy of solid tumors were ineffective, due to the actions of immunological checkpoints that prevented immune responses to self and to the tumor. Cytokine-inducible SH2-containing protein (CIS) is found to accumulate in NK cells in response to IL-15, which acts as a critical intracellular immune checkpoint that restricts NK cell function [28]. In particular, there is an inhibition of NK cell action by the interaction of killer-cell immunoglobulin-like receptors (KIRs) and the inhibitory NK cell receptor NKG2A on NK cells with major histocompatibility complex (MHC) class I molecules on tumor cells [29,30]. There is a parallel lack of specificity of CD8⁺ T cells stimulated by IL-15 due to the induction of SOCS3 in CD4⁺ helper T cells, thereby yielding ‘helpless’ CD8⁺ T cells [31]. To circumvent these checkpoints, the future development of IL-15 as a therapeutic will demand that it be used in combination therapy with antitumor agents increasing their efficacy. We have initiated such combination therapeutic clinical trials that include IL-15 with intralesional agonistic anti-CD40 to yield tumor specific CD8⁺ T cells, IL-15 administered with the immune checkpoint inhibitors anti-CTLA-4 and anti-PD-L1, and especially IL-15 with diverse tumor-directed monoclonal antibodies to increase their antibody-dependent cellular cytotoxicity (ADCC) and anticancer efficacy. Combination of IL-15 with other agents, such as the inhibitors of other immune checkpoints and anti-angiogenic molecules, may also require careful considerations [32,33]. It is hoped that with the use of these combination therapies IL-15 will soon play a prominent role in the treatment of patients with metastatic malignancies.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This work was supported in part by the Intramural Research Programs of the Center for Cancer Research at the National Cancer Institute [to TA Waldmann].