Phase II trials involving IFN-α were conducted by the National Cancer Institute in non-Hodgkin lymphoma (NHL) patients. Conflicting results were seen in regards to the impact of IFN-α induction monotherapy and maintenance, and when combined with chemotherapy, on survival in NHL patients. IFN-α was used in treating low grade indolent NHLs where it showed some activity, however complete response (CR) and overall response rates were only 10% and 48% respectively (26–28). The introduction of rituximab in the late 1990s, led to better results when it was combined with interferon due to enhanced ADCC (28). IFN-α may also have a role in the treatment of myeloproliferative diseases (29). Several strategies to overcome the narrow therapeutic index of IFN-α include delivering IFN-α into tumor cells via immunocytokines and genetically engineered dendritic cells expressing vectors encoding IFN-α.

Several studies established that continuous low dose of IL-2 results in expansion of NK cells while pulse intermediate dose IL-2 increases cytotoxic activity of NK cells (35–38). Synergistic activity against NHL was seen in mouse model when daily low dose IL-2 was administered with intermittent pulse intermediate dose IL2 and rituximab, likely due to enhanced ADCC mediated by NK cells. Although some early phase 1 studies showed promising results for the combination of rituximab and IL-2 (39, 40), other studies showed no significant clinical benefit (41, 42). Most of these studies used low dose IL-2, which could have preferentially expanded Tregs. The IL-2-based immunocytokines that showed positive results in preclinical lymphoma studies include L19-IL2 in NHL, HI-Leu 16-IL2 in lymphoma and HRS3scFv-IL12-Fc-IL12 in Hodgkin lymphoma. Diphteria toxin-IL-2 fused proteins showed promising results in a phase III trials of cutaneous T cell lymphoma (CTCL) patients (43). Currently, a phase II trial is ongoing for evaluating a similar diphtheria toxin-IL-2 fused protein with high bioavailability in patients with refractory/relapsed CTCL and peripheral T cell lymphomas (44).

High dose IL-2 is approved by the FDA for treatment of metastatic RCC and metastatic melanoma. However due to the various challenges with high dose IL-2 monotherapy including its substantial toxicity and modest efficacy, it is infrequently used and has largely been replaced by other immunotherapeutic agents (45). Low dose IL-2 could preferentially expand Tregs over the activation of NK and CD8 T cells (30). The high doses of IL-2 required for immune stimulation of NK and CD8 T cells is associated with serious side effects such as hypotension, organ failure, cytopenias and vascular leak syndrome (46). To improve the pharmacokinetics and pharmacodynamics of IL-2 and reduce its systemic toxicity several strategies have been pursued. Immunocytokines, in which IL-2 is linked to an antibody that targets tumor associated-antigens, have been efficacious in preclinical models and are currently being used in IL-2 based therapy trials (45). IL-2-based immunocytokines have been tested in combination with other cytokines which enhance the activation of NK cells (47). Combining IL-2 immunocytokines with other immunocytokines, chemotherapeutics, and immunotherapeutic agents such as immune checkpoint inhibitors will likely be implemented in future IL-2-based immunotherapies (45, 48). The administration of intratumoral IL-2 may decrease systemic toxicity (45, 48). Adjusting the design of manufactured IL-2/IL-2 immunocytokines to preferentially stimulate NK and CD8 T cells over the expansion of Tregs is one of approaches to optimize IL-2-based therapies. The structure of IL-2 cytokine/immunocytokines have been engineered to have higher affinity for the IL-2Rβγc heterodimer and diminished binding to the IL-2Rα on Tregs. IL-2 has also been successfully combined with an IL-2 antibody that masks the IL-2Rα binding site of IL-2, thus abolishing binding of engineered IL-2 to Tregs. Pegylation of IL-2-based immunocytokines can increase their half life and also block IL-2 from binding to IL-2Rα (49–51). Novel IL-2 fusion proteins have been designed to enhance the activity and proliferation of NK cells and have shown promising results in preclinical models. The OCMP-mutIL-2 fusion protein binds with high affinity to NKG2D on NK cells and contains mutations that confer preferential binding to IL-2Rβγc rather than IL2Rα (52, 53). Adjusting the structure of the IL-2-based therapies can also alleviate the risk of toxicities such as vascular leak syndrome.

The decrease in function of Th1 cells and underproduction of IFN-γ in the tumor environment of CTCL, makes IL-12 an attractive therapeutic option for this lymphoma subtype, especially in as much as cutaneous lesions could be easily accessed (54, 60). In a study of 10 patients with CTCL who received biweekly intralesional or subcutaneous (SC) injections of IL-12 monotherapy, 9 patients were available for response assessment at safety analysis time, responses occurred in 8/9 patients including 2 CR. Side effects were short lived (15). SC IL-12 monotherapy was also evaluated in a phase II trial of 23 patients with early stage mycosis fungoides whom received at least 3 prior antilymphoma therapies, 10/23 patients received IL-12 injections for at least 6 months and continued for 2 years. 73% had partial and minor responses while 22% had stable disease. However, 52% eventually progressed but some still achieved responses when continuing IL-12. 5 patients stopped therapy due to side effects and 1 patient died with hemolytic anemia that was likely attributed to the drug (16). A recent phase 2 study (NCT02542124) combining IL-12 with low dose total skin electron beam therapy has been initiated with the rationale that IL-12 may also decrease radiation toxicity.

The first study of IL-12 monotherapy in NHL and Hodgkin lymphoma (HL) patients was reported by Younes et al. (17) in which 32 patients with NHL and 10 patients with HL who received a median of three prior antilymphoma therapies were enrolled. NHL patients had diffuse large B-cell lymphoma (DLBCL) and grade 1-2 follicular lymphoma (FL). 11 patients received 250 ng/kg of intravenous (IV) IL-12 daily for 5 days every 3 weeks (preceded by an initial test dose of 250 ng/kg) and 31 patients received twice-weekly IL-12 SC injections at 500 ng/kg (in case of toxicity, the dosage was reduced to 300 ng/kg). 39 patients (93%) were present for response assessment. 6/29 patients (21%) with NHL had partial remission (PR) or CR while 10 patients (34%) had stable disease. FL patients had better responses and lower rate of progression than DLBCL patients (27% vs 64%). Median progression free survival (PFS) for indolent NHL and aggressive NHL patients was 6 and 2 months respectively. Overall responses in those who received IV and SC IL-12 were 40% and 7% respectively. All responders had low burden disease. IL-12 did not affect the CD4 T cell count but the mean CD8 T cell count was significantly increased. Levels of VEGF and bFGF (reflecting angiogenesis) were either decreased or stable after IL-12 injections in most patients who responded, and were stable or increased in most patients who did not respond or progressed. No responses were seen in patients with HL but 5/10 patients (50%) had stable disease with median PFS of 2 months. All HL patients received SC IL-12 and it is possible this could have influenced the results. IL-12 was tolerable in most patients with 95% developing fevers that mostly responding to acetaminophen. Fatigue, malaise and arthralgia were commonly reported and were more common in those who received IV IL-12. 9 patients required dose reductions including 3 patients who developed grade 3 hepatotoxicity and all received SC IL-12. All hepatotoxicities were grade 1 and 2. Treatment was prematurely stopped in 4 patients (17). In a phase 1 study of a group with heterogenous types of lymphomas, IL-12 was given SC twice weekly at doses 500 and 300 ng/kg or lower along with rituximab with the aim to enhance NK mediated ADCC. PR and CR were seen in 69% with more responses in those who received higher doses of IL-12 (18). However, in the phase 2 study, comparable response rates were seen with rituximab monotherapy and combination of rituximab and IL-12 (19). A phase II study has been initiated to evaluate the combination of IL-12 with salvage R-ICE or R-DHAP in patients with relapsed/refractory aggressive B-cell NHL with the rationale that IL-12 may enhance responses but also decrease chemotherapy toxicity (NCT02544724).

The safety of using IV IL-12 after autologous peripheral blood stem cell transplantation (PBSCT) was established in a phase 1 trial by Robertson et al. in which 12 patients including 10 with hematologic malignancies (8 NHL, 2 HL, 2 plasma cell myeloma) with very high risk of relapse received three different initial doses of IV IL-12. Time to progression ranged from 10.5 to 50.8 months with 3/8 of the NHL patients and 1/2 of HL patients, did not progress during the observation period (median of 32.4 months after initiating IL-12 therapy) (61). An obstacle to optimal IL-12-based immunotherapy after PBSCT is the profound, acquired STAT4 deficiency that can occur in this setting (62).

Significant systemic toxicities have been reported in studies with IL-12 that are mainly due to increased production of IFN-γ, TNF, IP-10 and MIG. Reported adverse events and toxicities in clinical trials include fatigue, fever, arthralgia, headaches and malaise (63–67). Erythropoietin analogues and G-CSF were successful in reducing anemia and bone marrow suppression respectively (54). It seems that systemic toxicity may be reduced if IL-12 is delivered locally into the tumor bed and is being implemented in current IL-12 clinical trials (54).

The anticancer effect of IL-12 in preclinical models was enhanced by combining it with chemotherapeutics, other cytokines, antibodies, antiangiogenic agents, radiotherapy, adoptive therapy, and tumor vaccines. For example, in a preclinical Burkitt lymphoma model, the anticancer effect of IL-12 was potentiated by the addition of vasostatin. Chemotherapeutic agents could augment the anticancer activity by increasing the release and presentation of tumor antigens. However, some chemotherapeutic agents could suppress the immune response by inhibiting immune effector cells (62).

Despite the anticancer activity seen in preclinical models, IL-12 monotherapy did not yield the same results in human clinical trials. One possible explanation for this discrepancy is the negative feedback and immunoregulatory effect that could result from increased IL-10 and TIM3 production that could lead to an increase in IFN-γ levels and increase PDL1 expression on cancer cells (54). Also, the human tumor environment may be more heterogenous than preclinical models with more immunosuppressive effect and tumor escape mechanisms.

Delivering IL-12 regionally or locally into the tumor environment may overcome its narrow therapeutic index by enhancing efficacy and decreasing systemic toxicity. With the advancement in genetic engineering techniques and adoptive cell therapies, IL-12 is regaining attention (68). In recent cancer experiments and early studies, IL-12 has been genetically delivered via engineered viruses or nonviral vectors such as IL-12 expressing plasmids (69–71). IL-12 plasmids are being studied in CTCL (54). Adoptive immunotherapy with the utilization of genetically modified IL-12 expressing lymphocytes have also been implemented (72). The idea of delivering CAR T cells transduced with inducible IL-12 is also being considered (73). IL-12 has also been combined with cancer vaccines including dendritic cell-tumor cell fusion vaccines (74). IL-12-based Immunocytokines in which IL-12 is attached to monoclonal antibodies that target specific tumor antigens may also help decrease systemic toxicity and improve efficacy of IL-12.

Despite the potent antitumor activity of IL-15 seen in some preclinical models, clinical results have not been as impressive, with issues that include a short half-life, relatively modest bioactivity in vivo, and reliance on transpresentation. Several approaches have been taken to overcome these challenges. Constructing superagonist complexes by combining IL-15 with various forms of IL-15Rα led to a superior antitumor activity than IL-15 monotherapy. Administration of a superagonist is thought to mimic transpresentation of IL-15 by IL-15Rα to effector cells expressing the IL-2/IL-15βγc heterodimer and thus enhancing IL-15 bioactivity (77, 79). Novel IL-15-based therapies have combined it in preclinical studies with other immunotherapeutic agents, such as monoclonal antibodies (rituximab, alemtuzumab), CD40 agents, cancer vaccines, and other cytokines (IL-12, IL-18, and IL-21) (82, 84). When these preclinical model approaches translated to human clinical trials, tumor responses have been disappointing, perhaps due to the immunosuppressive tumor microenvironment. Thus, combining IL-15-based therapies with immune checkpoint inhibitors that target the immunosuppressive environment is rational (82). IL-15-based therapy with dendritic vaccines and oncolytic viruses expressing IL-15 or IL-15- IL-15Rα are also being investigated in preclinical studies (82, 85). Toxicities and adverse events that were observed with recombinant IL-15 include fever, hypotension, appetite changes, weight loss, diarrhea, rash, transient grade 3–4 neutropenia and anemia. These adverse events were less severe when IL-15 was given SC and intermittently. A concern with the use of IL-15 arises from its potent stimulatory effect on NK cells and CD8 T cells, which could lead to autoimmune toxicities (78). Future studies are needed to determine the optimal administration route and dosage for IL-15- based therapy to minimize toxicity and enhance its bioactivity. Most of the ongoing/recruiting early phase clinical trials of IL-15 based therapy in lymphoma patients are using IL-15 super agonists and in combination with other anticancer therapeutics due to the challenges with IL-15 monotherapy mentioned before. Phase 1 trial has been initiated to evaluate the safety and efficacy of the IL-15 superagonist ALT -803 in combination with rituximab in patients with relapsed/refractory indolent NHL who received at least 1 prior rituximab treatment (NCT02384954).

IL-18 showed in vitro and in vivo synergistic activity when combined with the CD20 monoclonal antibody rituximab in a preclinical study (89). This led to a phase 1 study in which 19 patients with CD20+ NHL received rituximab and escalating doses of human recombinant IL-18. Eleven patients had rituximab-refractory disease. 5 patients achieved responses including 2 CR. The response rate was 62% for the 8 patients with indolent lymphoma who received doses of IL-18 that were associated with optimal biologic responses as defined by ancillary biomarker studies. Overall response rates were similar in patients with rituximab-refractory and rituximab-sensitive disease. The combination was deemed to be safe with a similar safety profile to IL-18 monotherapy. The objective responses are likely attributable to enhanced ADCC mediated by NK cells and monocyte/macrophages against rituximab-sensitized lymphoma cells (20). Although rituximab has been combined in clinical studies with other cytokines such as IL-2, IL-12 and GM-CSF, IL-18 has the advantage of enhancing the ADCC mediated by both NK cells and macrophages/monocytes. IL-2 and IL-12 enhance preferentially NK cell-mediated ADCC and GM-CSF preferentially enhances ADCC mediated by macrophages/monocytes (20). Ofatumumab is monoclonal antibody that binds to a different epitope of CD20 than that recognized by rituximab. Preclinical studies showed that IL-18 plus ofatumumab was more effective than IL-18 plus rituximab in a lymphoma xenograft model. IL-18-based immunotherapy might circumvent the acquired STAT4 deficiency that impairs IL-12-based treatment after PBSCT. A phase 1 study showed that the combination of ofatumumab with IL-18 in 9 patients with NHL (7 DLBCL) post high dose chemotherapy and PBSCT is feasible and tolerable with no dose limiting toxicities. Although efficacy could not be determined from this phase 1 study, the seven patients who were not taken off study early for lymphoma progression remained alive without evidence of active lymphoma at a median of 3.5 years post PBSCT. All of these patients have remained progression free for more than 2.4 years post PBSCT (21). Other approaches include utilizing IL-18 secreting CAR T cells and IL-18-based adoptive transfer (such as IL-12/IL-15/IL18 preactivated NK cells) (90, 91) with ongoing clinical trials as promising results were seen in early clinical trials of AML patients (92, 93).

IL-18 has a very favorable toxicity profile but has had limited efficacy as a monotherapy. IL-18 binding protein (IL-18BP) is a secreted antagonist that binds to IL-18 with high affinity and neutralizes its biological activity. IFN-γ induced by IL-18 may exert a negative feedback loop by increasing the expression of IL-18BP (91). Zhou et al. developed a form of IL-18 that is resistant to inhibition by IL-18BP and showed a potent antitumor responses in mice (94). IFN-γ induced by IL-18 could also stimulate upregulation of PD-L1 on tumors cells, providing a rationale for combining IL-18 and PD-1 checkpoint inhibitors in cancer immunotherapy. Other approaches to enhance the antitumor activity of IL-18 include combining it with chemotherapy and other immune therapeutics (monoclonal antibodies,other cytokines, cancer vaccines) (91). IL-18 secreting CAR T cells and IL-18-IL2 fusion proteins are under investigation.

Complicating the development of IL-18-based cancer immunotherapy are preclinical studies showing the IL-18 can promote tumor invasiveness and progression in some experiments. It is possible that IL-18 levels may have an impact on tumor regression/growth as administration of low dose IL-18 in murine melanoma model promoted metastasis due to suppressed mature NK cell number, while high doses of IL-18 that resulted in IL-18 serum levels > 1 ng/mL inhibited tumor growth without a decrease in mature NK cell number (95). The lowest doses of hIL-18 given to lymphoma patients resulted in IL-18 plasma levels of > 10 ng/mL and biomarker studies have shown in vivo activation rather than suppression of NK cells during IL-18-based immunotherapy (20, 87). Further studies are needed to determine the precise conditions under which IL-18 promotes or inhibits tumor growth versus regression. IL-18BP-based therapy may be a potential therapeutic modality for clinical scenarios where IL-18 can promote tumor growth.

The effect of IL-21 on B-cells is influenced by the stage differentiation of the B-lymphocyte. IL-21 can either enhance the proliferation or apoptosis of the B cells. Given that IL-21 has an indirect antitumor activity by enhancing the cytotoxic activity and proliferation of CD8 T cells and NK cells and that it can promote apoptosis of the B-cell at certain stages by upregulation of pro-apoptotic proteins such as Bim and downregulation of anti-apoptotic proteins, there has been an interest in studying IL-21 in the treatment of lymphoma. IL-21 led to complete regression of DLBCL in xenograft models and extended their survival by promoting apoptosis of B-cells (96, 97). Glebert et al. showed that IL-21 had antiproliferative effect on 2 mantle cell lymphoma (MCL) cell lines with increased pro-apoptotic proteins and decreased anti-apoptotic protein (98). IL-21 combined with rituximab was evaluated in a phase 1 clinical trial involving 21 patients with indolent R/R NHL including 9 patients with FL and 1 patient with marginal zone lymphoma. Of 19 patients with evaluable responses, objective response occurred in 8 patients including 3 CR/CRu and 5 PR. A third of the patients who had rituximab-resistant disease responded (22). Contrarily, clinical studies showed that IL-21 can promote the proliferation of Hodgkin lymphoma (99), EBV transformed lymphomas (100) and ALK-positive anaplastic large cell lymphoma (101) suggesting that blocking the effect of IL-21 may be a potential therapeutic approaches in these lymphoma subtypes.