Recent Advances in Immunotherapy in Metastatic NSCLC

Non-small cell lung cancer (NSCLC) is one of most common malignancies and the leading cause of cancer deaths worldwide. Despite advances in targeted therapies, majority of NSCLC patients do not have targetable genomic alterations. Nevertheless, recent discovery that NSCLC is an immunogenic tumor type, and several breakthroughs in immunotherapies have led to rapid expansion of this new treatment modality in NSCLC with recent FDA approvals of programed death receptor-1 inhibitors, such as nivolumab and pembrolizumab. Here, we review promising immunotherapeutic approaches in metastatic NSCLC, including checkpoint inhibitors, agents with other mechanisms of action, and immunotherapy combinations with other drugs. With advent of immunotherapy, therapeutic options in metastatic NSCLC are rapidly expanding with the hope to further expand life expectancy in metastatic lung cancer.

In 1957, Sir Macfarlane Burnet and Lewis Thomas first recognized the antigenicity of tumors and discovered immune surveillance in tumorigenesis (1). However, translation of the use of immune system for clinically meaningful benefits has started to realize only recently.
Compared to the circulating tumor antigen-specific T-cells, T-cells in the tumor microenvironment have impaired effector functions (2)(3)(4). Mice infected with chronic lymphocytic choriomeningitis virus (LCMV) show upregulation of negative costimulatory receptors on T-cells, including cytotoxic T-lymphocyte antigen 4 (CTLA-4), programed death receptor-1 (PD-1), T-cell immunoglobulin domain and mucin domain-3 (TIM-3), lymphocyte activation gene-3 protein (LAG-3), and others (5). The loss of T-cells effector function during chronic infections is similar in cancer where T-cells chronically exposed to tumor antigens enter a state of exhaustion (6). Major efforts concentrating on reversing T-cell exhaustion by inhibiting negative checkpoints with antibodies against PD-1 and its ligand, programed death receptor-1 ligand (PD-L1) are at the forefront of immunotherapy research.
Clinical success with immunotherapy was first seen in highly immunogenic malignant melanoma. High-dose interleukin-2 (IL-2) showed an objective response rate (ORR) of 16% (7). Interestingly, many complete responders experience long-term responses, and IL-2 was used extensively in melanoma patients despite high toxicities (8). This changed with approvals of CTLA-4 inhibitor ipilimumab in 2011, and PD-1 inhibitors, such as pembrolizumab and nivolumab, in 2014, which Recent Advances in Immunotherapy in Metastatic NSCLC Frontiers in Oncology | www.frontiersin.org November 2016 | Volume 6 | Article 239 have better side effect profiles. The response rates with ipilimumab are ~10%, while PD-1 inhibitors have ~40% ORR (8,9). Combination studies of ipilimumab and nivolumab have shown ORR of 42-60% (9,10), unprecedented in melanoma, have led to survival benefits and safer treatment options. Lung cancer has been viewed as non-immunogenic tumor where immunotherapies, such as BCG and IL-2, showed no efficacy (11,12). However, with recent success of immune checkpoint inhibitors, this has changed, and here we review the recent data in non-small cell lung cancer (NSCLC) and emerging immunotherapies.
eSTABLiSHeD SinGLe-AGenT CHeCKPOinT inHiBiTORS nivolumab Nivolumab is a fully human IgG4 antibody against PD-1, which can activate host immune system. In phase I studies in pretreated NSCLC patients, nivolumab showed ORR of 17-18% (13,14). A phase III trial in patients with advanced squamous NSCLC compared nivolumab with docetaxel, demonstrating median OS 9.2 vs. 6.2 months (15). In another phase III trial, patients with advanced non-squamous NSCLC progressing on platinumdoublet chemotherapy were randomized to receive nivolumab or docetaxel, showing median OS 12.2 months with nivolumab and 9.4 months with docetaxel (16). The subgroup of patients without numerical progression-free survival (PFS) advantage with nivolumab included never smokers, patients with KRAS wild type, EGFR-mutant tumors, although this study was not powered for subgroup analysis. PD-L1 expression was detected by IHC using a human PDL-1 antibody (Epitomics). Cutoffs were divided into three groups: 1, 5, and 10%. Quantifiable PDL-1 expression was seen in 78% patients. OS difference favoring nivolumab was found to be significant across all subgroups. No meaningful advantage in survival was seen in patients with PDL-1-negative tumors (16). FDA approved nivolumab for NSCLC patients progressing on platinum-doublet chemotherapy in 2015.
The role of nivolumab in patients with Eastern Cooperative Oncology Group performance status 2, as maintenance therapy, and in combination with other therapies remains to be established. A five-arm, phase II basket study has been launched with ~2000 NSCLC patients including PS2 patients, maintenance cohorts, and epidermal growth factor receptor positive (EGFR+)/ anaplasplastic lymphoma kinase positive (ALK+) patients treated with erlotinib or crizotinib with and without nivolumab (CHECKMATE 370; NCT02574078).

Durvalumab
Similar to atezolizumab, durvalumab is an IgG1 antibody to PD-L1. In a phase I study, durvalumab was administered to patients with NSCLC progressing after chemotherapy. Among 149 patients, ORR was 14% favoring PD-L1+ patients (ORR 23%) and squamous (ORR 21%) compared to non-squamous histology (ORR 10%) (23). In treatment-naive patients with advanced NSCLC, ORR was 25% in a recent phase I/II study (24). ATLANTIC is an ongoing phase II study with durvalumab in NSCLC patients progressing after at least two prior chemotherapies (NCT02087423).

PD-1, PD-L1 Targeting Agents: Selecting the Right Patient Population
Scagliotti et al. first demonstrated difference in survival in NSCLC patients treated with platinum-doublet chemotherapy based on histology, a biomarker of chemotherapy response. Squamous histology patients did better with cisplatin/gemcitabine, while adenocarcinoma patients had improved OS on cisplatin/pemetrexed combination (26).
Predicting responses to immunotherapy is more complicated; PD-L1 expression is one potential marker. PD-L1 expression can change over time. PD-L1 expression at initial tumor biopsy does not always correlate with antitumor activity (27). Thommen et al. evaluated expression of checkpoints on CD8+ T-cells from patients with advanced NSCLC and developed a cumulative inhibitory score based on sum expression of several receptors. Significant increase in the score was seen in patients with node positive, advanced stage, while tumor size did not affect it. An inverse correlation between PDL-1 expression and effector function of CD8+ T-cells treated with PD-1 antibody was observed (28). While PD-L1 expression is one of the markers for response in NSCLC, prior smoking and high mutational load correlate with responsiveness, suggesting that better predictive biomarkers or scores are needed (29). Schumacher et al. developed cancer immunogram with seven different immune parameters including mutational load to help predict responsiveness of tumors to checkpoint inhibitors (30). Efforts should be focused in this direction to help identifying the target population and to drive down treatment costs.

ipilimumab and Tremelimumab
Ipilimumab and tremelimumab are recombinant monoclonal antibodies against cytotoxic CTLA-4 that block CTLA-4 interaction with its ligands, such as CD80 and CD86. This blockade augments T-cell activation and proliferation, leading to tumor infiltration by T-cells and tumor regression (31). Although response rates with single-agent CTLA-4 inhibitor in NSCLC have been disappointing, there is considerable interest in exploiting this pathway in combination with other checkpoint inhibitors (32). In a phase II study, ipilimumab in combination with chemotherapy showed modest improvement in PFS in patients with advanced NSCLC (33). The benefit was limited to patients who received phased ipilimumab, allowing exposure to chemotherapy before ipilimumab (33). Encouraging results seen in early stage combination trials of CTLA-4 with PD-1 and PD-L1 inhibitors are discussed below.

epigenetic Approaches to improve immunotherapy Responses
Epigenetic silencing and loss of gene expression is a hallmark of cancer (34). DNA-demethylating and histone-deacetylating agents show modest benefit in solid tumors (35,36). Some patients may have durable response to subsequent chemotherapy or immunotherapy after receiving epigenetic modulators (37). Epigenetic therapy could sensitize tumors to subsequent therapy by altering tumor genomics, increasing tumor antigenicity (35,38,39). In a phase I/II trial, heavily pretreated NSCLC patients received AZA and etinostat. Among 65 patients treated in phase II portion, 2 responded (37). Six patients from this trial subsequently received PD-1 inhibitors and five of the six patients showed PR, including three patients with durable responses for 14-26 months (40).

Lymphocyte Activation Gene-3
Lymphocyte activation gene-3 is a transmembrane protein, closely related to CD4 that binds major histocompatibility complex II (42). LAG-3 is co-negative regulator of effector T cells (43).
Regulatory T-cells (Tregs) express LAG-3 that is upregulated in the presence of effector T-cells, leading to decreased antitumor immunity (44). LAG-3 antibodies inhibit Tregs (44). Preclinical data in multiple tumors demonstrated prolonged survival with dual PD-1/LAG-3 inhibition (45). Tissue samples from patients with ovarian epithelial cancers tumor-derived T-lymphocytes showed enrichment for coexpression of LAG-3 and PD-1 (46). Most of LAG-3 data at this time remain preclinical, and efficacy in solid malignancies remains to be seen. Urelumab (BMS-986016) and LAG525 are anti-LAG-3 antibodies investigated in early phase clinical trials in solid tumors including NSCLC alone and in combination with nivolumab or PDR-001 (NCT02658981, NCT01968109, NCT02460224).

eMeRGinG SinGLe-AGenT iMMUnOMODULATinG DRUGS Bavituximab
Bavituximab is a chimeric immunoglobulin antibody targeting a common phospholipid, phosphatidyl serine (PS), on tumor vascular endothelium (47). In normal cells, PS is expressed on internal membrane surface; however, in tumors, it becomes exposed on endothelial cells. Bavituximab binds to PS in a β2-glycoprotein I-dependent manner leading to antitumor effects such as antibody-dependent cell-mediated toxicity. Bavituximab prevents interaction between PS on apoptotic tumor cells and macrophages and decreases secretion of antiinflammatory immunosuppressive molecules, such as TGF-β and IL-10 (48,49).
In a phase II trial, docetaxel + bavituximab (D + B) was compared to docetaxel + placebo (D + P) in patients with locally advanced lung adenocarcinoma. The D + B combination was well tolerated, and results were encouraging with an improvement of ORR from 11.3% (D + P arm) to 17.1% (D + B) (50). The confirmatory phase III SUNRISE trial was halted when at 33% events it failed to show OS advantage of docetaxel + bavituximab combination (NCT01999673). A combination immunotherapy trial with durvalumab (PD-L1 inhibitor) and bavituximab in recurrent NSCLC is ongoing (NCT02673814).

indoleamine 2,3-Deoxygenase
Indoleamine 2,3-deoxygenase (IDO) is a cytoplasmic tryptophan catabolic enzyme that plays a significant role in T-cell suppression (51). High levels of IDO were first reported in human placenta (52). Inhibition of IDO in mice placenta leads to T-cell-mediated rejection of allogeneic conceptii (53). In addition to promoting immune tolerance to fetus in placenta, IDO-mediated depletion of tryptophan prevents excessive T-cell activation and lung inflammation (54). IDO supports development of immune tolerance to tumor antigens, activation of regulatory T-cells, and inhibition of cytotoxic T-cells and NK cells and is a relevant target in tumor immunology (55).
An ongoing trial is evaluating a combination of IDO inhibitor indoximod with docetaxel and Tergenpumatucel-L in NSCLC (NCT02460367). Ongoing basket trials in solid malignancies including NSCLC are testing other IDO inhibitors alone or in combination with checkpoint inhibitors (NCT02048709, NCT02073123).

eMeRGinG PD1/PDL1 iMMUnOTHeRAPY COMBinATiOnS wiTH CTLA4 inHiBiTORS, CHeMOTHeRAPY, AnD TARGeTeD THeRAPY
Immunotherapy combinations hold promise to further increase ORR and OS and outperform chemotherapy in NSCLC in the first line settings, based on success of nivolumab/ipilimumab combination in metastatic melanoma, now FDA approved based on CheckMate067 trial results (56).
Multiple studies are evaluating the role of frontline immunotherapy in treatment-naïve NSCLC patients. CheckMate012 trial included 148 treatment-naïve NSCLC patients with PDL-1 expression >1%; 80% patients had non-squamous histology. Arm A patients received nivolumab and ipilimumab Q3 weeks with ORR 13%, arm B -nivolumab 1 mg/kg Q2 weeks, ipilimumab 1 mg/kg Q6 weeks, ORR was 25%. Arm C had the best ORR of 39% when nivolumab was given 3 mg/ kg Q2 weeks and ipilimumab 1 mg/kg Q12 weeks. Arm D had an ORR 31% -nivolumab Q2 weeks, ipilimumab Q6 weeks. Median duration of response was not reached (57). A phase II single-arm ipilimumab-nivolumab trial in metastatic NSCLC is underway (CheckMate 568, NCT02659059). Important ongoing phase III studies that might change the treatment landscape of NSCLC are KEYNOTE-024 comparing pembrolizumab to chemotherapy in PD-L1 + NSCLC, and CheckMate 026 with nivolumab, both in treatment-naïve advanced NSCLC patients.
CheckMate 227, a phase III ipilimumab-nivolumab study in treatment-naïve advanced NSCLC in comparison to chemotherapy, is ongoing (NCT02477826). Combination of PD-L1 inhibitor durvalumab and CTLA4 inhibitor tremelimumab was recently studied in a phase Ib study of 102 pretreated NSCLC patients. Significant dose-limiting toxicities were seen in durvalumab 20 mg/kg, tremelimumab 3 mg/kg cohort. Encouraging ORR of 23% irrespective of PD-L1 status was observed in patients treated with durvalumab 20 mg/kg and tremelimumab 1 mg/kg (24). MYSTIC and NEPTUNE are ongoing phase III studies comparing this combination to platinum-based chemotherapy in treatment-naïve NSCLC patients (NCT02453282, NCT02542293).
In EGFR or ALK-positive NSCLC, several early phase trials are ongoing, combining PD-1 and CTLA-4 checkpoint inhibitors with EGFR (erlotinib, afatinib, osimertinib) or ALK (crizotinib) tyrosine kinase inhibitors (TKIs) in treatment-naive and previously treated patients (NCT01998126, NCT02511184). TATTON is a multi-arm phase Ib trial of durvalumab with osimertinib. Dose escalation arm included EGFR TKI-naïve patients, while the expansion cohort included patients progressing on prior TKIs. Although the combination showed clinical activity, 38% rate of interstitial lung disease was reported with the combination, disproportionate to what is expected with each agent alone. The combination arm has been put on hold (62). A phase I feasibility trial of afatinib and pembrolizumb for patients with EGFR mutation-positive NSCLC progressing on erlotinib is recruiting (NCT02364609).

COnCLUSiOn
Advances in immunotherapy in metastatic NSCLC now include nivolumab and pembrolizumab that are standard in second line settings, where both agents significantly improve survival. The treatment paradigm changed with introduction of immunotherapy and is expected to start impacting first-line treatment approaches soon. Ongoing challenges include understanding mechanisms of resistance to immunotherapy drugs, development of biomarkers, and predictive scores for better patient selection. Table 1 summarizes ongoing immunotherapy trials in advanced NSCLC, and Figure 1 describes mechanisms of action of select agents. New immunotherapy drugs and novel immunotherapy combinations in frontline and recurrent settings in NSCLC remains one of the most exciting and rapidly evolving areas in oncology, with hope to increase the numbers of long-term survivors with stage IV lung cancer in the near future.

AUTHOR COnTRiBUTiOnS
First author -PB -prepared manuscript and table. Corresponding author -YB -prepared part of the manuscript and table and provided guidance to the first author in preparing the manuscript; also helped in editing the article and the figure. Co-author -DOprepared the figure. Co-author -GS -helped with review and vital modifications along with suggestions to improve the content. Co-author -GG -helped with review and edition along with suggestions to improve the content.

FUnDinG
The study was supported by NCI P30 CA118100 Core Grant Funding to UNM Comprehensive Cancer Center (PB, DO, GG, and YB), NCI P30 Core Grant CA016672 to MD Anderson Cancer Center (GS), and DOD career development award LC140074 (YB).