Combination systemic therapies with immune checkpoint inhibitors in biliary tract cancer: effective but not enough?

KEYWORDS:

• Immunotherapy
• cholangiocarcinoma
• biliary tract cancer
• pembrolizumab
• durvalumab

1. Introduction

Biliary tract cancer (BTC), including intrahepatic cholangiocarcinoma (iCCA), extrahepatic cholangiocarcinoma (eCCA), and gallbladder carcinoma (GBC), represents a heterogenous group of malignant tumors arising from the epithelium of bile ducts or gallbladder [1]. Recent years have registered that the incidence of BTC, which accounts approximately for 10–15% of all hepatobiliary malignancies, has progressively increased worldwide, and a proportion of cases ranging from 60 to 70% is diagnosed with advanced stage – locally advanced or metastatic disease [2]. Surgical resection represents the only potentially curative treatment option for BTC patients, but even following radical surgery with curative intent, the 5-year overall survival (OS) is only 20–35% [3]. Systemic treatment options for BTC are limited, and more than a decade after the practice-changing ABC-02 phase III trial, cisplatin-gemcitabine remains the standard first-line therapy for patients with metastatic disease [4]. Combination chemotherapy with cisplatin plus gemcitabine resulted in an overall response rate (ORR) of approximately 26% and a median progression-free survival (PFS) and overall survival (OS) of 8.1 months and 11.7 months, respectively [5]. However, most BTCs experience disease progression and 20–25% of patients are deemed eligible to second-line treatment [6]. The recently published, randomized, ABC-06 phase III trial compared FOLFOX plus active symptom control versus active symptom control alone in patients whose disease progressed on first-line therapy, reporting a statistically significant OS improvement (median OS of 6.2 months and 5.3 months, respectively) [7]. Similarly, a phase IIb trial from South Korea highlighted improved PFS for liposomal irinotecan, 5-fluorouracil, and leucovorin compared to 5-fluorouracil and leucovorin alone (median PFS of 7.1 months and 1.4 months, respectively) [8].

In addition, through better genetic analysis and understanding of molecular biomarkers, targeted therapies have been explored in BTC patients, including fibroblast growth factor receptor-2 (FGFR-2) and isocitrate dehydrogenase-1 (IDH-1) inhibitors [9,10]; however, these treatments are limited to a small proportion of subjects harboring these genomic alterations.

Immunotherapy with immune checkpoint inhibitors (ICIs) has been assessed in patients with metastatic BTC, reporting controversial clinical benefit in several trials investigating ICI monotherapy [11]. Due to the low ORR of BTC immunotherapy, recent years have seen the presentation and publication of trials assessing combinatorial treatments including ICIs and conventional therapies to improve efficacy of this approach. In the current Editorial, we provide a critical overview of clinical trials assessing immune-based combinations in metastatic BTC patients, discussing pros and cons of this approach in this setting.

2. Expert opinion

ICIs restore T-cell-mediated tumor cell killing and deplete regulatory T cells by blocking immune checkpoint molecules including programmed cell death protein 1 (PD-1), programmed cell death ligand 1 (PD-L1), and cytotoxic T-lymphocyte antigen 4 (CTLA-4) [12]. This peculiar mechanism of action has attracted a lot of attention from preclinical and clinical researchers worldwide, leading to landmark clinical trials in several hematological and solid tumors. Interestingly, upregulation of PD-1 and/or PD-L1 has been highlighted in BTC tumor tissues, supporting the exploration of ICIs in patients with metastatic BTC [13]. However, the limited ORR and acquired resistance to PD-1 and PD-L1 inhibitors monotherapy have represented an important obstacle to the clinical development of these agents in BTC [14].

The phase II KEYNOTE-158 and the phase Ib KEYNOTE-028 clinical trials assessed the anti-PD-1 agent pembrolizumab for previously treated BTC patients (Table 1) [15]; in the intention-to-treat population, KEYNOTE-158 reported a discouraging ORR of 5.8% in the intention-to-treat population, with median PFS and OS of 2.0 and 7.4 months, respectively [15]. Similarly, ORR, median PFS, and median OS were 13.0%, 1.8 months, and 5.7 months in KEYNOTE-028 [15]. Another ICI, the PD-1 inhibitor nivolumab, has been assessed as monotherapy in metastatic BTC in a single-group, multicenter phase II trial [16]. According to the results of this study published by Kim and colleagues, single-agent nivolumab showed partial response and stable disease in 10 and 27 out of 45 pretreated patients, respectively. Median PFS and OS were 3.68 and 14.24 months [16]. The PD-L1 inhibitor durvalumab was explored in a phase I basket trial including BTCs, reporting a median OS of 8.1 months (95% CI, 5.6–10.1) and a median duration of response of 9.7 months [17].

Table 1. Table summarizing main phase I–II clinical trials evaluating immunotherapy in biliary tract cancer patients Abbreviations: CisGem, cisplatin plus gemcitabine; CTLA-4, Cytotoxic T-Lymphocyte Antigen 4; DCR: disease control rate; NE: not estimable; ORR, overall response rate; mo: months; mOS, median overall survival; PD-1, programmed death 1, mPFS: median progression-free survival; PR: partial response.

NCT name	Phase	Setting	Treatment Arm	Agents description	Results
NCT02628067 (KEYNOTE-158) [15]	II	Second- or later-line	Pembrolizumab	Pembrolizumab: PD-1 inhibitor	ORR: 5.8% (2.1–12.1) mOS: 7.4 mo (5.5–9.6) mPFS: 2.0 mo (1.9–2.1)
NCT02054806 (KEYNOTE-028) [15]	Ib	Second- or later-line	Pembrolizumab	Pembrolizumab: PD-1 inhibitor	ORR: 13.0% (2.8–33.6) mOS: 5.7 mo (3.1–9.8) mPFS: 1.8 mo (1.4–3.7)
JapicC- TI-153098 [28]	I	Second- or later-line	Nivolumab	Nivolumab: PD-1 inhibitor	ORR: 3.3% (0.7–13.6) mOS: 5.2 mo (4.5–8.7) mPFS: 1.4 mo (1.4–1.4)
NCT02829918 [16]	II	Second- or later-line	Nivolumab	Nivolumab: PD-1 inhibitor	PR: 22% DCR: 59% mOS: 14.2 mo (5.98-not reached) mPFS: 3.7 mo (2.3–5.69)
NCT01938612 [17]	I	First- or later-line	Durvalumab	Durvalumab: PD-L1 inhibitor	DCR: 16.7% mOS: 8.1 mo (5.6–10.1)
NCT02923934 (CA209-538) [30]	II	First- or later-line	Nivolumab plus ipilimumab	Nivolumab: PD-1 inhibitor Ipilimumab: CTLA-4 inhibitor	ORR: 23% DCR: 44% mOS: 5.7 mo (2.7–11.9) mPFS: 2.9 mo (2.2–4.6)
NCT01938612 [17]	I	First- or later-line	Durvalumab plus tremelimumab	Durvalumab: PD-L1 inhibitor Tremelimumab: CTLA-4 inhibitor	DCR: 32.2% mOS: 10.1 mo (6.2–11.4)
NCT03311789 [29]	II	First-line	Nivolumab plus CisGem	Nivolumab: PD-1 inhibitor	ORR: 55.6% DCR: 92.6% mOS: 8.5 mo (5.0–12.5) mPFS: 6.1 mo (3.4–8.2)
JapicCTI- 153,098 [28]	I	First-line	Nivolumab plus CisGem	Nivolumab: PD-1 inhibitor	ORR: 36.7% mOS: 15.4 mo (11.8-NE) mPFS: 4.2 mo (2.8–5.6)

Based on these premises and these results, researchers have recently focused their attention on combination regimens with ICIs plus other anticancer agents to increase clinical outcomes in this patient population. For example, the addition of a CTLA-4 inhibitor to an anti-PD-1 or anti-PD-L1 agent has been reported to enhance efficacy in several tumor types (e.g. melanoma, renal cell carcinoma, non-small-cell lung cancer, etc.) [18–20]. CTLA-4 and PD-1 ligands block CD3/CD28-mediated glucose metabolism, inhibiting the upregulation of Akt activity through several mechanisms; moreover, CTLA-4 inhibitors impair the process of T cell activation in peripheral lymph nodes, while PD-1 inhibitors improve the effector T cell function in the tumor microenvironment [21]. Thus, simultaneous targeting of these two immune checkpoints has been suggested to boost antitumor activity. Similarly, the combination of an ICI plus antiangiogenic agents represents another treatment regimen worth looking forward to, due to the potential synergistic effect of these drugs [22]. Angiogenesis is of landmark importance for primary tumor growth, having a close relationship with the immune system [23]; interestingly, antiangiogenic agents may prevent the formation of blood vessels in cancer and inhibit tumor cell growth. VEGF binding to VEGFR2 on effector T cells upregulates PD-1 expression in CD+ T lymphocytes, impairing cell proliferation [24]; thus, antiangiogenic therapy combined with immunotherapy may have a synergistic effect, as shown in hepatocellular carcinoma, renal cell carcinoma, lung cancer, and other solid tumors [25–27].

As regards the field of BTC, the PD-1 inhibitor nivolumab has been tested in combination with the reference doublet cisplatin – gemcitabine as front-line treatment for metastatic BTC patients, highlighting a median OS of 15.4 months and a median PFS of 4.2 months, with 11 out of 30 patients showing an objective response [28]. In another phase II trial investigating nivolumab – gemcitabine – cisplatin, the combination therapy showed an ORR of 55.6%, with a median PFS of 6.1 months and a median OS of 8.5 months [29]. Durvalumab has been investigated in combination with the CTLA-4 inhibitor tremelimumab in a phase I trial, where the immune-based combination showed a median OS of 10.1 months (95% CI, 6.2–11.4) [17]. Similarly, a phase II trial investigated the role of the double checkpoint blockade with nivolumab plus the anti-CTLA-4 agent ipilimumab in 39 BTC patients, observing an ORR and DCR of 23% and 44%, respectively [30]. Median OS and PFS were 5.7 months and 2.9 months in BTC patients treated with nivolumab – ipilimumab [30].

Some weeks ago, the results of the phase III TOPAZ-1 have been presented, following a press release stating that the combination of the reference doublet cisplatin – gemcitabine plus durvalumab reported clinically meaningful and statistically significant survival benefit compared to chemotherapy alone in treatment-naïve BTC patients [31]. At the first interim analysis, the addition of the PD-L1 inhibitor durvalumab conferred a 20% reduction in the risk of death compared with gemcitabine and cisplatin alone, thus meeting the primary end point of the study (Hazard Ratio [HR], 0.80; 95% Confidence Interval [CI], 0.66–0.97; p = 0.021). In addition, PFS and ORR were also improved with chemoimmunotherapy versus chemotherapy alone. Interestingly, separation in the OS curves for the two treatment arms did not occur until approximately 6 months after treatment initiation, with these data being consistent with several previous immunotherapy clinical trials. Based on the promising TOPAZ-1 results, durvalumab in combination with gemcitabine – cisplatin has the potential to represent a novel front-line standard in BTCs [31].

Finding specific biomarkers and screening out populations, which could advantage from immunotherapy before treatment, is key to achieve precise treatment. At present, several tissue-based biomarkers have been tested as predictors of response, including PD-L1 status, tumor mutational burden (TMB), microsatellite instability (MSI), mismatch repair (MMR) gene defects, and tumor infiltrating lymphocytes (TILs) [32–34]. However, available evidence does not support the use of these biomarkers. For example, the KEYNOTE-028 and KEYNOTE-158 basket trials reported no association between response to ICIs and PD-L1 expression, while longer median PFS and higher ORR were observed in PD-L1 positive BTC patients receiving nivolumab in the study by Kim et al. [15,16].

As previously reported, ICI monotherapy has modest antitumor activity in BTC patients, something that underscores the urgent need for clinical trials in which immunotherapy is combined with other anticancer agents. Combined therapy has the potential to play a particularly important role in improving the effectiveness of ICIs, expanding the beneficiary population and overcoming drug resistance, especially following the encouraging results of TOPAZ-1. Several ongoing studies have the potential to add further information regarding the potential role of immunotherapy plus chemotherapy, as in the case of the KEYNOTE-966 study assessing first-line pembrolizumab plus gemcitabine – cisplatin. Similarly, the combination of ICIs with anti-VEGF agents is supported by a strong biological rationale: hypoxia, resulting from altered blood supply, supports malignant cell escape from immune surveillance, impairing the function of immune effector cells. Moreover, the activation of hypoxia-inducible factor 1 alpha has been associated with the upregulation of PD-L1 expression. Antiangiogenic agents further enhance hypoxia, and since PD-L1 is preferably expressed in hypoxic areas, anti-VEGF agents can improve T-cell activity. In the case of the PD-L1 inhibitor atezolizumab and the VEGF inhibitor bevacizumab, for example, these two agents display synergistic activity, with the latter presenting immunomodulatory effects. Thus, these combinations could also play a role in BTC, especially in some specific subgroups of patients. At the same time, the promising results of TOPAZ-1 may prompt clinicians and researchers to consider the expansion of ICIs to the earlier stage of the disease. More efforts are needed to better understand how the addition of immunotherapy such as durvalumab to standard chemotherapy could impact and which BTC patients could benefit from these treatments.

Declaration of interests

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.

Declaration of interest

A reviewer of this manuscript received consulting fees from Agios, Amgen, Bayer, BMS, HalioDx, Incyte, Merck Serono, MSD, Pierre Fabre Oncologie, Roche, Sanofi, Servier, and Viatris. Honoraria for lectures: Amgen, Bayer, BMS, Merck Serono, MSD, Pierre Fabre Oncologie, Roche, Sanofi, Servier, and Viatris. They also received travel expenses from Amgen, Bayer, BMS, Merck Serono, MSD, Pierre Fabre Oncologie, Roche, Sanofi, and Servier. Peer reviewers of this manuscript have no other relevant financial relationships or otherwise to disclose.

Additional information

Funding

This paper was not funded.