Role of Indoleamine-2,3-Dioxygenase Inhibitors in Salvage Therapy for Non-Muscle Invasive Bladder Cancer

Due to significant risks of cancer recurrence and progression, and limited options after intravesical Bacillus Calmette Guerin (BCG) therapy, there is a critical unmet need to identify novel treatments for those patients with BCG-unresponsive bladder cancer. There is active investigation of immunotherapies which provide both biologic and clinical rationales for indoleamine-2,3- dioxygenase inhibitors in salvage therapy for non-muscle invasive bladder cancer.

Key points

•

The role of immunotherapy in the treatment of bladder cancer has spurred investigation into regulatory pathways of the immune system and subsequent interactions with tumor biology.
•

Patients with BCG unresponsive non-muscle invasive bladder cancer have an unmet need for bladder sparing options, and existing therapies will require combination approaches to overcome immune escape.
•

Indoleamine 2,3-dioxgenase 1 (IDO1) inhibitors are a class of small molecule drugs which have demonstrated efficacy in combination with existing immunotherapies in multiple types of advanced cancer, with increasing application in metastatic, muscle-invasive, and non-muscle invasive bladder cancer.
•

Manipulating the immune microenvironment with IDO1 inhibitors can increase sensitivity to existing therapies and augment the host response.

Introduction to indoleamine-2,3-dioxygenase 1 and role in immunosuppression

Indoleamine-2,3-dioxygenase 1 (IDO1) is an intracellular immunoregulatory enzyme that serves as the first and rate-limiting step for the conversion of tryptophan to kynurenine ( Fig. 1 A). The enzyme is highly expressed in the placenta and mucosa of the female genitourinary tract, lungs, and lymphoid organs, and its role has been implicated in maternal tolerance to fetus, allograft protection, and cancer progression. ^, This heme-dependent enzyme inserts an oxygen molecule across the 2 to 3 bond of the indole moiety of tryptophan. Early observations in the 1950s showed elevated tryptophan catabolism in patients with bladder cancer, implicating the role of IDO1 in oncogenesis. Subsequent studies have demonstrated that IDO1 mediates host immunosuppression due to the sensitivity of T cells to the presence of kynurenine and tryptophan deprivation, thereby driving immune escape. Of note, a distinct pathway for the conversion of tryptophan to kynurenine by Trp 2,3-dioxygenase (TDO) also has been described, normally present in the liver and also found to have increased activity in tumor microenvironments. ^,

IDO1 activity is upregulated in tumor, stromal, and innate immune cells, where its expression is linked with shifting the tumor microenvironment from immune activity to immune tolerance. Several cancer cell types constitutively express IDO1, including melanoma, non-small cell cancer, renal cell carcinoma, bladder cancer, and cervical cancer. Levels of kynurenine and kynurenine to tryptophan ratios are elevated in the serum of patients with cancer compared with healthy subjects. By depleting tryptophan, IDO1 facilitates the suppression of CD8+ T effector and natural killer cells, generation and activation of CD4+ T regulatory and myeloid-derived suppressor cells, and promotion of tumor angiogenesis ^, ( Fig. 1 B). This mechanism occurs in concert with programmed cell death protein 1 (PD-1) pathways to induce immune quiescence. In fact, IDO1 expression increases after PD-1 blockade in patients with melanoma and renal cell carcinoma, suggesting a potential resistance mechanism of tumor cells to avoid immune detection. ^, In patients with bladder cancer failing Bacillus Calmette-Guerin (BCG) treatment, PD-L1 (ligand of PD-1) expression is associated with higher-grade tumors, infiltration by mononuclear cells, and BCG-induced granulomata. In parallel, IDO1 expression is correlated with tumor size, grade, and stage in non–muscle invasive bladder cancer (NMIBC), and is present in 62% of bladder cancers. ^, By altering the tryptophan to kynurenine ratios, IDO1 can promote the tumor microenvironment. Downstream effector pathways of tryptophan depletion include repression of the mammalian target of rapamycin (mTOR) pathway and eukaryotic initiation factor 2 (eIF2) activation. Increased kynurenine levels also activate the proinflammatory aryl-hydrocarbon receptor linked to carcinogenesis. ^, ^, Thus, by altering the tryptophan to kynurenine ratio and by depleting tryptophan, IDO1 can trigger a robust response used by tumor cells to avoid immune surveillance.

Development of indoleamine-2,3-dioxygenase 1 inhibitors in clinical trials

Several factors support IDO1 and tryptophan catabolic pathways as promising therapeutic targets in bladder cancer. First, IDO1 is a small-molecule, single-chain catalytic enzyme with a well-defined biochemistry. Second, serum tryptophan measurements allow monitoring of IDO1 inhibition, making it an attractive target for drug development. Interest in the IDO1 enzyme as a target for cancer therapy began more than a decade ago, when preclinical models inhibiting IDO1 enhanced the efficacy of chemotherapy, radiotherapy, and immune checkpoint therapy without a significant increase in adverse events (AEs).

In the past 5 years, 3 IDO1 inhibitors have reached at least phase II clinical trials and have demonstrated preliminary efficacy in patients with multiple types of advanced cancer. These IDO inhibitors are indoximod, epacadostat, and linrodostat mesylate (linrodostat; BMS-986205). Other inhibitors are being tested in phase I trials, and additional patents have been filed (Roche, Merck) in the preclinical space ( Tables 1 and 2 ).

Table 1

Current state of indoleamine-2,3-dioxygenase 1 inhibitors in development

Data from Prendergast GC, Malachowski WP, DuHadaway JB, Muller AJ. Discovery of IDO1 Inhibitors: From Bench to Bedside. Cancer Res. 2017;77(24):6795-6811. https://doi-org.easyaccess1.lib.cuhk.edu.hk/10.1158/0008-5472.CAN-17-2285 ; and Gyulveszi G, Fischer C, Mirolo M, et al. Abstract LB-085: RG70099: A novel, highly potent dual IDO1/TDO inhibitor to reverse metabolic suppression of immune cells in the tumor micro-environment. Cancer Res . 2016;76(14 Supplement):LB-085-LB-085. https://doi-org.easyaccess1.lib.cuhk.edu.hk/10.1158/1538-7445.AM2016-LB-085 .

Drug	Company	Mechanism	Trial Start	Program Status
Indoximod NLG-8186	NewLink	mTORC1 downstream activator	2008	Phase III
Epacadostat INCB024360	Incyte	Catalytic inhibitor	2012	Phase III
Linrodostat BMS 986205	BMS	Catalytic inhibitor	2015	Phase III
Navoximod NLG-919	NewLink	Catalytic inhibitor	2015	Phase IB
PF-06840003	Pfizer	Catalytic inhibitor	2016	Phase I
IOM2983	Merck	Unknown	Unknown	Preclinical
RG-70099	Roche	Dual IDO/TDO inhibitor	Unknown	Preclinical

Table 2

Existing data on IDO1 inhibitors in patients with advanced bladder cancer

Drug	Patients, n	Objective Response Rate, %	Disease Control Rate, %	Clinical Trial	Phase	Comment
Linrodostat + nivolumab	27	37	56	CA017-003	I/IIa	Anti-PD-1 monotherapy historically 15%–20% response rate
Epacadostat + pembrolizumab	40	35	53	ECHO-202/ KEYNOTE-037	I/II	Higher PD-L1 associated with higher overall response rate (64% vs 13%)

Abbreviations: IDO1, indoleamine-2,3-dioxygenase 1; PD-1, programmed cell death protein 1; PD-L1, ligand of PD-1.

Indoximod

Indoximod inhibits mTORC1, a downstream effector of IDO1, and has progressed to phase III clinical trials. In an earlier phase II trial of 102 patients with heavily pretreated melanoma, indoximod with pembrolizumab achieved an overall response rate (ORR, including partial and complete responses) of 56% (19% complete response), which exceeds the 33% ORR of pembrolizumab monotherapy. Given the inherent limitations of cross-trial comparisons between a single-arm phase II study, and a randomized controlled phase III study, and the limited sample size of this phase II trial compared with the much larger global phase III trial of nivolumab and ipilimumab in metastatic melanoma (CheckMate-067), these response rates appear to be comparable to the currently approved regimen of ipilimumab plus nivolumab but with potentially less immune-related toxicity. In another phase II trial, patients with metastatic castrate-resistant prostate cancer treated with indoximod versus placebo after sipuleucel-T experienced greater than twofold increase in radiographic progression-free survival. Other studies in breast cancer and acute myeloid leukemia are ongoing. NCT01792050 is a phase II clinical trial randomizing docetaxel plus indoximod versus docetaxel alone in patients with metastatic breast cancer with the primary endpoint of progression-free survival; this trial is actively recruiting. NCT02835729 is a phase II trial also in active recruitment, randomizing patients with acute myeloid leukemia to either indoximod or placebo while undergoing induction therapy with cytarabine and idarubicin; however, indoximod has not yet been studied in bladder cancer.

Epacadostat

Epacadostat is a selective inhibitor against IDO1 that has been studied in several phase II trials. ECHO-202 was a multi-disease cohort of patients with advanced solid tumors treated with pembrolizumab and epacadostat, showing impressive early response rates of 33% to 58% across multiple tumor types (non-small cell lung cancer, renal cell carcinoma, endometrial adenocarcinoma, urothelial carcinoma, and squamous cell carcinoma of the head and neck), compared with 16% to 33% with pembrolizumab monotherapy. By RECIST v1.1 criteria, 8 of 62 patients achieved complete response (5 patients with treatment-naïve melanoma, 3 patients with advanced and pretreated melanoma, urothelial cancer, and endometrial adenocarcinoma). Seventeen patients achieved partial response (treatment-naïve melanoma in 6 patients, non–small-cell lung cancer in 5 patients, and 2 patients each with renal cell carcinoma and urothelial carcinoma, and 1 patient each with endometrial adenocarcinoma, and head and neck cancer). Twenty-four percent of patients reported grade 3/4 AEs. ECHO-204 was a phase II study examining epacadostat in combination with nivolumab across multiple disease cohorts. Results included promising efficacy data with a 62% ORR across all patients, and a 65% ORR across 40 treatment-naïve patients. Grade 3 or higher treatment-related adverse events were reported in 48% of patients treated at an epacadostat dose of 300 mg twice a day (most common AEs being rash and alanine aminotransferase increase), but in only 13% of patients treated at an epacadostat dose of 100 mg twice a day, which was the dosage that was selected for subsequent phase III studies. In melanoma, epacadostat in combination with anti-PD-1 agents (pembrolizumab or nivolumab) achieved rates of response and disease control comparable to the approved combination of PD-1 and CTLA-4 antibodies but without similar rates of immune-mediated AEs and in a manner independent of PD-L1 levels. Similarly promising results were found in patients with triple-negative breast, head and neck, endometrial, lung, renal, and urothelial cancers. ^, These findings were believed to be proof of concept that addition of epacadostat to checkpoint inhibitors could bolster response rates by synergizing with existing pharmacologic pathways.

Despite promising phase II data, a subsequent phase III trial in melanoma (ECHO-301/KEYNOTE-252) showed no improvement in progression-free survival with the epacadostat/pembrolizumab combination when compared with pembrolizumab alone. In 706 patients with unresectable or metastatic melanoma randomized to pembrolizumab with either epacadostat or placebo, there was no difference in median progression-free survival (4.7 vs 4.9 months, hazard ratio [HR] 1.0; 37% for both groups at 12 months), and overall survival was not expected to reach statistical significance based on results of interim analysis (HR 1.13; confidence interval 0.86–1.49; P = .807; 74% in both groups at 12 months). Median follow-up was 14 months and 72.5% of patients were PD-L1 positive. It remains unclear whether the negative results from this trial were related to lack of efficacy, inadequate dosing of epacadostat, patient selection independent of biomarkers, or persistence of a parallel TDO pathway that mitigated the effects of the IDO1 inhibitor. As previously discussed, the TDO pathway has similar yet distinct downstream effects that may also deplete tryptophan and increase kynurenine. In melanoma, for example, IDO1 suppression alone may be insufficient to relieve the immunosuppressive effect of kynurenine. The findings of ECHO-301/KEYNOTE-252 triggered the early termination of several other studies of IDO1 inhibitors, including linrodostat in melanoma and head and neck cancer (Bristol Meyers Squibb), indoximod in metastatic melanoma (Indigo 301, NewLink), and epacadostat (Incyte) in 6 other late-stage clinical trials.

Linrodostat (BMS-986205): application in bladder cancer

Linrodostat is a selective, potent, once-daily, oral IDO1 inhibitor that occupies the heme cofactor binding site, preventing activation of the IDO1 pathway to reduce kynurenine production. Despite recent development challenges for IDO1 inhibitors, including the negative ECHO-301/KEYNOTE-252 study, the role of IDO inhibition in combination with anti-PD-L1 therapy in the treatment of bladder cancer appears promising. Phase II data from the combinations of pembrolizumab with epacadostat and nivolumab with linrodostat showed that in patients with advanced urothelial carcinoma who had received prior systemic treatment (cisplatinum or alternative), epacadostat and linrodostat potentially conferred additional antitumor activity when combined with checkpoint inhibitors compared with checkpoint inhibitors alone (ECHO-202/KEYNOTE-037 and CA017-003 studies, respectively). In ECHO-202, a total of 40 patients with advanced bladder cancer refractory to prior platinum-based therapy demonstrated a 35% ORR (all partial responders) with a disease control rate (DCR: complete response, partial response, stable disease) of 57%. Progression-free survival and biomarker analyses are ongoing. The most common AEs were fatigue, rash, and increased amylase, and grade 3 or higher AEs occurred in 20% of patients.

In the phase I/IIa dose escalation study CA017-003, linrodostat in combination with nivolumab demonstrated clinical activity in 30 patients with advanced bladder cancer, all of whom had received at least 1 line of prior therapy. Eighty-three percent of these patients had baseline visceral metastases, including one-third with liver metastasis, and 50% had PD-L1–positive tumors. The study demonstrated that the 100-mg dose of linrodostat was better tolerated than the 200-mg dose with similar efficacy. A 37% ORR and 56% DCR was observed in 27 patients with advanced bladder cancer who had not previously received immunotherapy. Median time to response was 7.8 weeks. The response rate was higher in PD-L1 positive versus PD-L1 negative tumors (50% vs 30% ORR, 64 vs 50% DCR). ORR was 30% in patients with baseline visceral metastases and 33% in patients with baseline liver metastases. Deep and durable reduction in tumor burden was observed in both PD-L1–positive and PD-L1–negative tumors, with some responses lasting beyond treatment discontinuation. Grade 3/4 treatment-related AEs occurred in 37% of all patients: 21% in patients treated with the 100-mg dose and 50% in patients treated with the 200-mg dose of linrodostat. Consequently, the 100-mg daily dose of linrodostat was established as the selected dose for phase II and III clinical trials moving forward.

These results have laid the groundwork for 2 pivotal trials with linrodostat in bladder cancer. The phase II trial (CheckMate 9UT; NCT03519256 ) will investigate 4 different treatment regimens (nivolumab alone, nivolumab plus BCG, nivolumab plus linrodostat, or nivolumab plus linrodostat and BCG) in BCG-unresponsive, high-risk NMIBC ( Fig. 2 ). Recruitment for this study is ongoing with an estimated goal of 480 enrolled patients. Follow-up will continue until disease recurrence/progression or for 5 years, and will include routine surveillance cystoscopy, cytology, and biopsy per American Urological Association and European Association of Urology guidelines. Primary endpoints include complete response rate (patients with carcinoma in situ (CIS)) and event-free survival (in all other patients). Secondary endpoints include progression-free survival as well as the safety and tolerability of the investigational treatments. Pharmacokinetics, potential predictive biomarkers, and changes in patient-reported outcome for quality of life also will be assessed.