In the ongoing search for novel immunotherapies that might rival or surpass the efficacy of immune checkpoint inhibitors, drugs targeting IDO1—a key enzyme in tryptophan metabolism—have been a major focus in recent years.
In the ongoing search for novel immunotherapies that might rival or surpass the efficacy of immune checkpoint inhibitors (ICIs), drugs targeting IDO1—a key enzyme in tryptophan metabolism—have been a major focus in recent years.
Lead candidate epacadostat seemed to offer significant potential in combination with the PD-1—targeting ICI pembrolizumab (Keytruda), and Incyte Corporation advanced the regimen rapidly into phase III clinical trials.1
Other pharmaceutical companies also were diving into the field. Another immunotherapy success story appeared imminent until the failure of the ECHO-301 trial testing the combination of epacadostat and pembrolizumab2 sent shockwaves through the field.
Following these negative results, Incyte took a sledgehammer to its IDO inhibitor program, halting enrollment across a swathe of ongoing pivotal trials.3 The anxiety spread. Bristol- Myers Squibb undertook a complete overhaul of its IDO inhibitor program,4-6 and it was the final nail in the coffin for a tentative IDO collaboration between Genentech and NewLink Genetics, an alliance that had suffered setbacks of its own in recent years.7,8
The ECHO-301 trial may serve as a cautionary tale against rushing into inadequately planned clinical testing of an agent with limited and suboptimal preclinical data. Some investigators have argued that these caveats should be considered “when weighing the implications of this one trial against the overall potential” of agents that inhibit IDO and TDO, another tryptophan-catabolizing enzyme. The rapid pace of clinical development may have come at the cost of a thorough understanding of this highly complex pathway and its nuanced role in cancer development.9
These difficulties, however, have not deterred new entrants into the field. Several companies are developing IDO1 inhibitors with a novel mechanism of action.10-12
Dual inhibitors of IDO1 and TDO are also being developed to address functional redundancy between these enzymes, a potential mechanism of resistance to IDO1 inhibition. Other studies are exploring alternative ways of targeting the broader pathway of tryptophan metabolism that IDO1 regulates.9
Figure. The Role of IDO1 in the Tumor Microenvironment17 (Click to Enlarge)
Leveraging an Essential Amino Acid
Tryptophan is the rarest of the essential amino acids—those that must be acquired through the diet—and, as such, its metabolism is tightly regulated.13 Several biochemical pathways are involved in tryptophan breakdown, but the kynurenine (KYN) pathway predominates, converting tryptophan into biologically active metabolites, including the eponymous KYN.13-17
The rate-limiting step in this pathway involves several heme-containing metalloenzymes: IDO1, or indoleamine 2,3-dioxygenase 1; the closely related IDO2; and TDO, or tryptophan 2,3-dioxygenase. All 3 enzymes catalyze the same reaction, but in different tissue types. TDO is most highly expressed in the liver and is the major mediator of hepatic tryptophan metabolism. IDO1 has a broader range of expression than TDO and IDO2 and recognizes other indole-containing substrates besides tryptophan. IDO2 also is much less well studied than the other enzymes.13-18
IDO and TDO expression is regulated by a range of nutritional and inflammatory signals. TDO can be activated by tryptophan, cholesterol, prostaglandin E2, and others, while regulators of IDO activity include interferon gamma, interleukin 6, and tumor necrosis factor alpha.13-17
IDO-mediated tryptophan depletion has 3 major downstream effects. First, it activates general control nonderepressible 2 (GCN2), a serine/threonine kinase that senses amino acid deficiency and phosphorylates eukaryotic translation initiation factor 2 alpha, leading to reduced protein production and inducing apoptosis of effector T (Teff) cells.13-17
Second, IDO-mediated tryptophan metabolism inhibits a master regulator of metabolism, mTOR, which feeds into a network of amino acid sensors, indicating to the cell that the available supply of tryptophan is low. Third and last, it activates the aryl hydrocarbon receptor (AhR), a transcription factor that controls the function of a plethora of immune cells. Adding to the complexity, AhR itself can activate IDO1, both directly and indirectly, establishing a positive regulatory feedback loop13-17 (Figure17).
IDO’s Immunosuppressive Role
Increased tryptophan metabolism in advanced cancers was first noted in the 1950s,19 but this remained a relatively obscure observation until IDO was linked to immunosuppression, suggesting that cancer cells were hijacking this immunosuppressive activity to evade immune detection.13,18
IDO1 overexpression has now been observed across many tumor types.20,21 It has been found to be under the control of BIN1, a tumor suppressor protein that is commonly attenuated in cancer, revealing one of the mechanisms of increased IDO1 expression in tumors.13,17
Moreover, IDO1 is expressed not only by cancer cells but also by stromal, endothelial, and immune cells of the tumor microenvironment. IDO2 and TDO have also been shown to be overexpressed in some cancers, and IDO1 and TDO may be co-expressed in a substantial proportion of tumors.9,13,22
IDO1 expression has varied effects on different immune cells, including blocking the activation of Teff cells, stimulating the activation of regulatory T cells, and inhibiting natural killer cell function, in addition to promoting the differentiation of tolerogenic dendritic cells and the activation and expansion of myeloid-derived suppressor cells. Collectively, this fosters a highly immunosuppressive local environment.15,16,18
Table. IDO Inhibitors in Clinical Development (Click to Enlarge)
Development Ramps Up
Indoximod became the first IDO inhibitor to undergo clinical testing.
In a phase II study in 135 patients with metastatic pancreatic cancer, indoximod was combined with gemcitabine and nab-paclitaxel (Abraxane). Among 104 patients evaluable for efficacy, the overall response rate (ORR) was 46.2%, with a complete response (CR) rate of 1.0% and a partial response (PR) rate of 45.2%. The combination was well tolerated. Median overall survival was 10.9 months, but the study did not meet its prespecified goal of a 30% reduction in hazard ratio.24
A separate phase II trial of indoximod in combination with taxane chemotherapy (docetaxel or paclitaxel) also failed to meet its primary endpoint of statistically significant improvement in progression-free survival (PFS) in patients with metastatic breast cancer.8
Several bona fide direct catalytic inhibitors of IDO1 have been developed. The lead candidate, epacadostat, competes with tryptophan for binding to the IDO1 catalytic site. In concert with Genentech, NewLink Genetics developed navoximod, a tryptophan noncompetitive inhibitor that is also a weak inhibitor of TDO.16,23
Bristol-Myers Squibb and Pfizer also threw their hats into the ring with the development of linrodostat (BMS-986205) and PF-06840003, respectively.25,26
Clinical trials of epacadostat monotherapy were disappointing, with no objective responses.27-29 Studies suggesting that IDO1 overexpression may serve as a mechanism of resistance to ICIs, targeting PD-1 and its ligand PD-L1 and demonstrating synergy between the 2 types of therapy, served as the rationale for pursuing clinical trials of this combination, which initially showed great promise.16
Promising Double Act?
Enthusiasm for combining IDO inhibitors with ICIs began building after findings from a study of indoximod plus investigator’s choice of nivolumab (Opdivo), pembrolizumab, or ipilimumab (Yervoy) in patients with advanced melanoma produced an ORR of 55.7% and a CR rate of 18.6%.30
In the ECHO-202 study, epacadostat combined with pembrolizumab showed efficacy across several tumor types in patients with advanced cancers. Among 62 patients, the ORR was 40.3%, including 8 CRs and 17 PRs. Patients with melanoma had an ORR of 55%, and responses were also seen in patients with non—small cell lung cancer (NSCLC), renal cell carcinoma, endometrial adenocarcinoma, urothelial carcinoma, and head and neck squamous cell carcinoma (HNSCC).1
The combination of navoximod plus atezolizumab (Tecentriq) elicited PRs in 9% of patients and stable disease in a further 17% in a phase I clinical trial.31 In a phase I/II trial of linrodostat in combination with nivolumab, among 27 patients with immunotherapy-naïve advanced bladder cancer, the ORR was 37%, including 3 CRs and 7 PRs.32
On the basis of these promising data, numerous IDO inhibitor—ICI combinations advanced to phase III clinical trials. The future looked rosy until Incyte reported the findings from the ECHO-301 trial, which evaluated the combination of epacadostat and pembrolizumab.
A total of 706 patients with advanced melanoma were randomly assigned to receive either the combination or pembrolizumab plus placebo. Over a median follow-up of 12.4 months, across all prespecified and post hoc subgroups examined, there was no significant difference in PFS between the 2 groups (median PFS, 4.7 months vs 4.9 months, respectively). An independent data monitoring committee recommended that the study be stopped, and no additional efficacy analyses are planned at this time.2
The failure stunned the research community and threw up a large roadblock to IDO inhibitor development. It prompted Incyte to halt enrollment in ongoing pivotal trials of epacadostat paired with pembrolizumab, nivolumab, or durvalumab (Imfinzi).3 Bristol- Myers Squibb followed suit and suspended trials of linrodostat in NSCLC and HNSCC.4
NewLink Genetics halted the randomization portion of Indigo301, a phase I/II study involving combinations of indoximod with nivolumab or pembrolizumab in advanced melanoma.33 The company also shifted the focus for indoximod away from melanoma to 3 other indications: recurrent pediatric brain tumors, diffuse intrinsic pontine glioma, and acute myeloid leukemia (AML).5-7
The ECHO-301 failure was also the final straw for the collaboration between NewLink Genetics and Genentech, which was terminated in May 2018.7
Despite the domino effect triggered by ECHO-301 and other clinical trial failures, enthusiasm for IDO inhibitors has not been completely extinguished. At least 8 agents are under study in ongoing clinical trials, including select studies into epacadostat and indoximod that are still recruiting (Table).
New data from initial indoximod studies in brain tumors and AML have been promising. Among 29 pediatric patients with advanced brain tumors enrolled in an ongoing phase I clinical trial of indoximod combined with chemotherapy and radiation therapy, the median PFS was 6.2 months, and 9 patients were continuing treatment at the time of data presentation.34
In an ongoing phase I trial, among 25 patients with newly diagnosed AML treated with indoximod and induction chemotherapy who received at least 1 dose of indoximod, 84% achieved remission, and the rate of minimal residual disease—negative status was promising at 83%.35
NewLink Genetics is developing a prodrug of indoximod, NLG802, and early clinical trial results were presented in May at the Immuno-Oncology World Congress in Barcelona, Spain. NLG802 exhibited a tolerable safety profile with no dose-limiting toxicities in findings from a phase I study in 26 patients with recurrent advanced solid tumors refractory to chemotherapy or targeted agents.36
Meanwhile, Bristol-Myers Squibb is pursuing linrodostat across a range of tumor types in multiple clinical trials. These include a phase III trial that seeks to recruit 1200 patients with non—muscle-invasive bladder cancer. The study is testing neoadjuvant gemcitabine, cisplatin, and nivolumab with or without linrodostat, followed by nivolumab with or without linrodostat versus chemotherapy after radical cystectomy (NCT03661320).
Linrodostat is being evaluated in combination with nivolumab and nivolumab plus ipilimumab in an estimated 907 patients with advanced cancers, including melanoma and NSCLC (NCT02658890).
New entrants also are entering the field. Kyowa Hakko Kirin, a Japanese company, and Eli Lilly are both developing IDO inhibitors with a novel mechanism of action. Since many conventional IDO inhibitors have structural similarity to tryptophan, they can activate AhR downstream of IDO. This can trigger the positive feedback activation of IDO, potentially confounding the effects of IDO inhibition.14 The new IDO inhibitors target the apo form of IDO1, which lacks heme, and reportedly avoid inadvertent AhR agonism.10-12
Additionally, several pharmaceutical companies are exploring more potent dual IDO/TDO inhibitors, with HTI-1090 the first to enter clinical trials.
Finally, if AhR is the most important downstream effector in regard to IDO’s role in cancer immunosuppression, as some investigators suspect, then inhibiting this protein could also have significant anticancer efficacy.16 An AhR-inhibiting agent, BAY2416964, recently entered first-in-human clinical testing in patients with advanced solid tumors (NCT04069026).
The first downstream mediator of this immunosuppressive role to be identified was activation of GCN2, but the importance of the AhR and mTOR pathways has also come to light more recently. The jury is still out on which pathway is most important in this regard, and it may be that effector pathways have different degrees of importance in different tumor types.9,16,17,18,23
These efforts suggest that the IDO pathway story is far from over.