Immune checkpoint inhibitors in malignant pleural mesothelioma: promises and challenges

KEYWORDS:

• Malignant pleural mesothelioma
• immunotherapy
• immune-checkpoint inhibitors

Malignant pleural mesothelioma (MPM) is a tumor with a poor prognosis, whose incidence is increasing worldwide due to widespread previous exposure to asbestos. Only few patients are candidate to multimodality treatment including surgery, while most are treated with chemotherapy alone during the course of their disease. The combination of pemetrexed with a platinum compound has been the standard of care in the first-line setting for over a decade [1,2]. Recently, the addition of bevacizumab to pemetrexed plus cisplatin has shown to significantly improve overall survival (OS) in a large phase III randomized trial [3]. Unfortunately, nearly all MPM patients progress during or after first-line treatment. In MPM patients progressing after a pemetrexed-based regimen, there is no standard second-line therapy. Single agent chemotherapy with gemcitabine or vinorelbine has shown modest activity; in selected cases with a prolonged response to first-line platinum/pemetrexed, re-treatment with a pemetrexed-based regimen can be considered, with a weak level of evidence, based on small retrospective series. Most studies with targeted therapies have led to disappointing results [4]. Accordingly, there is an unmet need for effective therapy in the relapsed setting.

Recently, immunotherapy involving the targeting of immune checkpoints such as cytotoxic T-lymphocyte-associated protein 4 (CTLA-4) and programmed cell death 1 (PD-1) has shown impressive results in several cancers including melanoma, renal cell cancer, and lung cancer [5]. The emergency of these new therapies is going to change the whole cancer therapy scenario in the next few years. Based on this upcoming evidence, therapy with immune checkpoint inhibitors is being evaluated also in patients with unresectable MPM. The anti-CTLA-4 monoclonal antibody tremelimumab was evaluated in two small consecutive phase II trials, in which the drug was administered every 3 months [6] or with an intensified schedule [7]. In the first trial, 2 of the 29 treated patients achieved a long-lasting partial response (PR), 7 had a long-lasting stable disease (SD, median duration 12.4 months), for a disease control rate (DCR) of 31% [6]; moreover, landmark analysis at 1 and 2 years identified survival rates of 48.3% and 36.7%. These results were confirmed by the subsequent study [7], in which tremelimumab was administered at a dose of 10 mg/kg every 4 weeks for 6 cycles, followed by the same dose every 12 weeks until disease progression. In this trial, 4 of 29 patients achieved PR (14%) and 11 SD, for a DCR of 52%. Median OS was 11 months, and 1-year survival 52%. The most commonly reported toxicities were fever, and gastrointestinal and dermatological adverse events. These promising results have led to launch the DETERMINE trial (NCT01843374), a randomized, double-blind, placebo-controlled phase II–III study. The primary objective of this study was to compare the OS of subjects with unresectable malignant pleural or peritoneal mesothelioma treated in the second- and third-line setting with tremelimumab compared to placebo. Secondary objectives included comparisons of DCR, progression-free survival (PFS), patient-reported outcomes, overall response rate, duration of response, and safety. Unfortunately, the study failed to achieve its primary end point [8]. Tremelimumab is now being studied in combination with durvalumab, an anti-PD-Ligand 1 (PD-L1) agent, in a phase II trial in the first- and second-line setting ongoing at the University Hospital of Siena, Italy (NCT02588131). Preclinical data have suggested that targeting both PD-1/PD-L1 and CTLA-4 may have additive or synergistic effects. This has been reported also in clinical trials in other cancers, namely melanoma and renal cell cancer patients; the combination regimens, however, have been associated with increased immune-related toxicities [9,10].

PD-1 is a transmembrane inhibitory immunoreceptor expressed by activated T cells that negatively regulates immune responses by interaction with its ligand PD-L1, a member of the B7 gene family. Blockade of PD-1 or PD-L1 derepresses T-cell activation, unleashing a clinical immune response towards the tumor [5]. The expression of PD-L1 has been reported in up to 45% of mesothelioma samples in different series, with a higher rate in sarcomatoid MPM, and has been associated to poor prognosis [11]. Targeting of PD-1 axis in mesothelioma has demonstrated promising efficacy. In a phase IB multi-cohort study in PD-L1 positive solid tumors (KEYNOTE-028), an objective response rate of 28% and a DCR of 76% were observed in 25 MPM patients who received pembrolizumab, an anti-PD-1 antibody, at the dose of 10 mg/kg every 2 weeks [12]. Median PFS in this pretreated population was 5.8 months, with 50% of patients progression-free at 6 months. Safety profile was manageable, with grade 3–4 toxicities occurring in four patients (16%), namely fever, uveitis, hepatotoxicity, and thrombocytopenia. Interestingly, no relationship between level of PD-L1 expression on tumor and immune cells within tumor nests and frequency of response was observed. Based on these results, further evaluation of pembrolizumab was planned, with a single-center phase II trial ongoing at the University of Chicago (NCT02399371). In this trial, an initial cohort of unselected group of pleural or peritoneal mesothelioma patients will be treated; in case of observed activity, prospective enrollment will follow using a biomarker enrichment strategy for PD-L1. Furthermore, a similar phase II trial with nivolumab, another anti-PD-1 antibody, in patients with relapsed mesothelioma is ongoing at the Netherlands Cancer Institute (NCT02497508). Nivolumab is also under evaluation in a randomized phase II study sponsored by the IFCT (Intergroupe Francophone de Cancerologie Thoracique), comparing efficacy and safety of second- or third-line treatment with nivolumab monotherapy or nivolumab plus the anti-CTLA-4 monoclonal antibody ipilimumab (NCT02716272).

PD-L1 blockade has also demonstrated promising efficacy in patients with mesothelioma. In a phase IB study, 20 patients with unresectable pleural or peritoneal mesothelioma, progressing after prior platinum/pemetrexed-containing chemotherapy, were treated with avelumab, an anti-PD-L1 antibody, at the dose of 10 mg/kg every 2 weeks [13]. Patients were heavily pretreated, with a median of two prior treatments. A durable PR was observed in three (15%) patients; SD was observed in nine additional patients, for an overall DCR of 60%. Median PFS was 16.3 weeks, and the 12-week-PFS rate was 66.7%. The most common adverse events were infusion-related reactions, fatigue, fever, and pruritus. Severe toxicities included colitis, decreased lymphocyte count, and increased blood creatine phosphokinase. Several important points need to be clarified regarding the use of PD-1/PD-L1 checkpoint inhibitors in MPM and, generally, in solid tumors. Maybe the most important issue is the need of a predictive biomarker driving patient selection. PD-L1 has been implicated as a putative biomarker of efficacy [14]; however, the association between PD-L1 expression and efficacy remains controversial, mainly because of the use of different companion tests and different cutoffs of positivity in the development of the various drugs. In mesothelioma, the pembrolizumab KEYNOTE-028 study failed to demonstrate a clear association between the level of PD-L1 expression and treatment efficacy [12]. However, criteria for PD-L1 positivity at study entry were quite large (membranous expression in ≥1% of tumor and associated inflammatory cells or positive staining in stroma), and the analysis was conducted in a very small sample size. In the avelumab study, MPM patients were not selected based on PD-L1 positivity [13]. In other cancers, mutation burden has been associated with increased neo-antigen load and sensitivity to PD-1 blockade [15]. A subset of mesotheliomas exhibits significant genomic instability and high mutational burden [16]; therefore, mutational load is being explored as a further marker of sensitivity to PD-1/PD-L1 inhibitors. Another important issue with immune checkpoint inhibitors is their toxicity. These agents generate a new burden of immune-related adverse events (irAEs), which remain largely unknown to the broad oncology community. The main irAEs include diarrhea, colitis, hepatitis, skin toxicities, and endocrinopathies such as hypophysitis and thyroid dysfunction. Although severe irAEs remain rare (especially under single agent therapy), they can become life threatening if not anticipated and managed appropriately [17].

In conclusion, immune checkpoint inhibition seems a promising therapeutic strategy in MPM. However, results are still very preliminary, and most trials are still ongoing. The CTLA-4 inhibitor tremelimumab administered as a single agent has failed to improve patient survival in the second- and third-line setting. Phase II trials evaluating PD-1 and PD-L1 inhibitors are ongoing, with early promising results. Combinations with targeting of both CTLA-4 and PD-1 axis are also being explored. Further biomarker studies are required to identify subgroups of good-responder patients, in order to optimize the benefit and the cost-effectiveness of these drugs in MPM. Besides the efficacy profile, a proper management of short- and medium-term adverse events will be important in improving quality of life and long-term outcomes.

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.