Using Immune Checkpoint Inhibitors in Lung Cancer
Using Immune Checkpoint Inhibitors in Lung Cancer
Immune checkpoint inhibition using targeted monoclonal antibodies is changing the treatment paradigm for lung cancer. Approval by the US Food and Drug Administration of two anti–programmed death 1 immune checkpoint inhibitors for second-line treatment of advanced or metastatic non–small-cell lung cancer (NSCLC) has led to increased use of these agents in the clinic. Ongoing clinical trials are evaluating the administration of immune checkpoint inhibitors alone or in combination with each other, and in combination with targeted therapy, radiation therapy, and chemotherapy regimens. Other trials are evaluating these monoclonal antibodies in small-cell lung cancer and across a variety of treatment regimens and disease stages in NSCLC. Ongoing translational work to identify relevant biomarkers will deepen our understanding of when and how to use immune checkpoint agents in our patients with lung cancer. We must continue to improve the cost-benefit ratios for efficacy, safety, and outcomes associated with the use of these medications. This is an exciting time in the field of lung cancer research as we work to understand how best to use this novel class of agents.
Immune checkpoint inhibition is changing the treatment paradigm for many cancer types by unleashing patients’ own immune systems against their tumors. Immune checkpoint inhibitors have been found to statistically improve the survival of patients with previously treated advanced or metastatic non–small-cell lung cancer (NSCLC).[1-4] Research into the use of immune checkpoint inhibitors in this setting led to rapid approval of these agents for the treatment of NSCLC by the US Food and Drug Administration (FDA) in 2015 (Table 1). This research, in turn, has led to the initiation of multiple studies to determine how best to optimize response and outcomes across a spectrum of combinatorial approaches, patient characteristics, and treatment lines in NSCLC (Table 1). Evaluation of potential patient, tumor, and biomarker features that can be used to predict likelihood of response and progression of resistance in lung cancer is ongoing, and is essential to treatment optimization. Concurrent with this expanded research, there is a need for further discussion of practical matters such as cost-benefit ratios, response assessment, and toxicity management as these newly approved monoclonal antibodies gain traction in the clinical setting.
Lung Cancer as an Immune Checkpoint Target
Historically, NSCLC was thought to be nonimmunogenic, as traditional immunotherapeutics such as interleukin-2 or interferon failed to generate clinical responses in late-stage clinical trials. The discovery of immune checkpoints—proteins in inhibitory pathways within the host immune system that function to induce self-tolerance and prevent autoimmunity, but that are exploited by tumor cells to evade immune recognition and response—led to the hypothesis that blockade of these co-receptors and pathways could induce a robust systemic antitumor response (Figure).[5-7] Currently the most well-elucidated inhibitory immune checkpoint co-receptor and pathway, respectively, are cytotoxic T-lymphocyte–associated antigen 4 (CTLA-4) and the programmed death 1 (PD-1)/programmed death ligand 1 (PD-L1) pathway. Multiple other immune checkpoints exist—including other inhibitory receptors, such as T-cell immunoglobulin and mucin domain–containing molecule-3 (TIM-3), lymphocyte activation gene 3 (LAG-3), B-lymphocyte and T-lymphocyte attenuator (BTLA), and V-domain immunoglobulin-containing suppressor of T-cell activation (VISTA); as well as activating receptors such as CD28, tumor necrosis factor receptor superfamily, member 4 (TNFRSF4 [also known as OX40 and CD134]), glucocorticoid-induced tumor necrosis factor receptor–related protein (GITR), CD137, CD27, and herpesvirus entry mediator (HVEM). Development of monoclonal antibodies that could interrupt mechanisms by which tumors evade the immune system evolved from research showing upregulation of CTLA-4 and the PD-1 pathway in multiple tumor types, including lung cancer.[6,9] Due to encouraging results observed with immune checkpoint antibodies across several tumor types, the use of these agents in the oncology clinic is rapidly expanding. Three immune checkpoint antibodies—the anti–CTLA-4 agent ipilimumab and the anti–PD-1 agents nivolumab and pembrolizumab—are currently FDA-approved for treatment of metastatic melanoma; nivolumab and pembrolizumab are approved for use in previously treated advanced or metastatic NSCLC; nivolumab is also approved for the treatment of previously treated advanced or metastatic renal cell carcinoma and previously treated classical Hodgkin lymphoma; and the anti–PD-L1 agent atezolizumab was most recently granted accelerated approval for previously treated advanced or metastatic urothelial carcinoma.
In the first phase I trials evaluating the use of PD-1/PD-L1 pathway immune checkpoint inhibitors, encouraging results were observed in the NSCLC patient cohorts. The first phase I study of single-agent therapy with nivolumab, a fully human IgG4 monoclonal antibody against PD-1, in refractory solid tumors revealed encouraging response rates and a tolerable safety profile. In the follow-up phase Ib study of nivolumab, the objective response rate (ORR) in the NSCLC patient cohort was 18%. It was also found to be well tolerated, with a 14% rate of grade 3/4 drug-related adverse events (AEs) across all tumor cohorts. Among all cohort responders followed for 1 year or longer, 64.5% of the responses were deemed durable in that they were ongoing for at least 1 year. Also in a phase I study, the anti–PD-L1 antibody BMS-936559 yielded an ORR of 10.2% in the NSCLC patient cohort. In this trial, much as in the phase Ib trial just mentioned, across all patients followed for at least 1 year, a response duration of at least 1 year was observed in 50% of responders, and the agent was well tolerated, with a 9% rate of grade 3/4 treatment-related AEs.
Published Results of Immune Checkpoint Inhibitor Clinical Trials in NSCLC
Subsequent reported results of trials assessing immune checkpoint inhibition in patients with NSCLC have revealed positive findings in favor of their use (Table 2). In the long-term follow-up period of a trial of nivolumab monotherapy, impressive results were reported for an expansion cohort of patients with previously treated advanced NSCLC. For all nivolumab doses, median overall survival (OS) was 9.9 months and the ORR was 17%; nivolumab at 3 mg/kg, the dose selected for further investigation, yielded a median OS of 14.9 months and an ORR of 24%. Perhaps most encouraging was the duration of response: at the 3-mg/kg dose, the 1-, 2-, and 3-year OS rates were 56%, 42%, and 27%, respectively—survival outcomes that were previously unheard of in this heavily pretreated patient population. For the 17% of patients who had objective responses, the estimated median duration of response was 17 months. The authors also noted that an additional 5% of patients had unconventional patterns of immune responses, in that the tumor kinetics showed a reduction in target lesions in the presence of new lesions, or regression of target lesions after initial growth. Another 10% of patients had stable disease lasting at least 6 months. This indicated that patients were receiving clinical benefit from nivolumab that was not necessarily included in the classic Response Evaluation Criteria in Solid Tumors (RECIST)-based response rate. Although the treatment was found to be well tolerated overall, with a 14% rate of grade 3/4 treatment-related AEs, three treatment-related deaths associated with pneumonitis were reported.
The phase II CheckMate 063 study evaluated nivolumab monotherapy in patients with advanced, refractory squamous NSCLC (n = 117). The ORR was 14.5%; however, because 77% of the responders had an ongoing response at the time of outcomes analysis, the median duration of response was not reached. Median OS was 8.2 months, and the 1-year OS rate was 40.8%. Again, treatment was well tolerated, with a 17% rate of grade 3/4 treatment-related AEs; this included a 3% rate of high-grade pneumonitis and two treatment-associated deaths (attributed to pneumonia and ischemic stroke).
The phase III CheckMate 017 trial evaluated nivolumab vs docetaxel in previously treated patients with advanced or metastatic squamous NSCLC (n = 272). The ORRs were 20% vs 9%, respectively. Survival outcomes were also statistically significantly improved: median OS was 9.2 months with nivolumab vs 6 months for patients treated with docetaxel. The 1-year OS rates were 42% vs 24%, and the rates of treatment-related grade 3/4 AEs were 7% vs 55% for nivolumab and docetaxel, respectively. CheckMate 057, a similar phase III trial, compared nivolumab vs docetaxel in patients with previously treated advanced or metastatic nonsquamous NSCLC (n = 582). Again, both ORR (19% vs 12%) and median OS (12.2 months vs 9.4 months) were significantly improved in the nivolumab arm. In addition, the 1-year OS rates were 51% in the nivolumab group vs 39% in the docetaxel group, and the rates of grade 3/4 treatment-related AEs were 10% with nivolumab vs 54% with docetaxel. One treatment-related death, attributed to encephalitis, occurred in the nivolumab arm. Interestingly, positive PD-L1 membrane staining was associated with greater treatment efficacy in the nonsquamous study population (CheckMate 057), but not in the squamous study population (CheckMate 017). Taken together with the other previously described study results, these investigations led to FDA approval of nivolumab monotherapy in previously treated advanced or metastatic NSCLC.
TO PUT THAT INTO CONTEXT
Edward B. Garon, MD, MS
David Geffen School of Medicine at UCLA Los Angeles, California
How Has the Role of Checkpoint Inhibitors in Lung Cancer Evolved?
The integration of inhibitors of the programmed death 1 (PD-1) checkpoint into the standard care of non–small-cell lung cancer (NSCLC) has been rapid, and new and important studies continue to be presented with incredible frequency. Since the publication in June 2012 of data on therapy with nivolumab for patients with solid malignancies, the US Food and Drug Administration has approved two PD-1 inhibitors for patients with previously treated NSCLC, and we now know that in a population selected based on high-level expression of programmed death ligand 1 (PD-L1), progression-free survival and overall survival are superior with PD-1 inhibition as opposed to standard chemotherapy.
What Are Some Important Considerations Regarding the Use of These Agents?
In their excellent review of the current state of immune checkpoint inhibition in lung cancer, Drs. Marrone and Brahmer focus their discussion on seven compounds, namely two PD-1 inhibitors, three PD-L1 inhibitors, and two cytotoxic T-lymphocyte–associated antigen 4 (CTLA-4) inhibitors. However, multiple additional inhibitors of these and other checkpoints are in development. Of note, no agent not targeting the PD-1/PD-L1 interaction has shown clear evidence of activity in NSCLC to date, making the benefit of combining multiple checkpoint inhibitors uncertain at this time. In addition to touching on the efficacy and toxicity of these agents, the authors address the very important consideration of cost, which is often overlooked in the literature about these agents. In a common malignancy such as lung cancer, it will be important to address these issues, particularly as we evaluate combinations of multiple costly immune checkpoint inhibitors.
Financial Disclosure: Dr. Garon’s institution has received research funding from AstraZeneca, Bristol-Myers Squibb, Genentech, Merck, and Pfizer.
The multi-arm phase I KEYNOTE-001 trial evaluated the humanized IgG4 kappa isotype PD-1 antibody pembrolizumab as monotherapy in patients with both systemic therapy–naive and treatment-refractory NSCLC (n = 495). Across all patients, the ORR was 19.4%, the median duration of response was 12.5 months, and the median OS was 12 months. Among previously untreated patients, the median OS was 16.2 months and the ORR was 24.8% (n = 101). Among previously treated patients, the median OS was 9.3 months and the ORR was 18% (n = 394). Grades 3 to 5 treatment-related AEs were reported in 9.5% of all patients, with one fatal case of pneumonitis. This study found that tumor PD-L1 status positivity, defined as samples with more than 50% of cells staining positive, was correlated with ORR (reported as 45.2%), median OS (not reached), and median progression-free survival (mPFS; 6.3 months).
The phase II/III KEYNOTE-010 trial evaluated pembrolizumab at two doses compared with docetaxel in previously treated patients with PD-L1–positive advanced or metastatic NSCLC. In this study, PD-L1 positivity was defined as tumors with more than 1% of cells staining positive. This study found a significant improvement in median OS among patients in both pembrolizumab arms (10.4 months at the 2-mg/kg dose and 12.7 months at the 10-mg/kg dose) compared with those in the docetaxel arm (8.5 months); there was no statistical improvement in PFS in the pembrolizumab arms. However, among the patients with more than 50% of tumor cells staining positive, both median OS (14.9 to 17.3 months vs 8.2 months with docetaxel) and PFS (5 to 5.2 months vs 4.1 months with docetaxel) were significantly improved in the pembrolizumab arms. Rates of grade 3–5 treatment-related AEs were less common with pembrolizumab (13% and 16% at the 2- and 10-mg/kg doses, respectively) than with docetaxel (35%). Together, these two studies, along with a companion diagnostic, the PD-L1 IHC 22C3 pharmDx test, led to the FDA approval of pembrolizumab as therapy for patients with previously treated PD-L1–positive advanced or metastatic NSCLC.
Two trials have evaluated atezolizumab (MPDL3280A), a humanized IgG1 anti–PD-L1 antibody, in previously treated advanced or metastatic NSCLC. The first dose-escalation and expansion phase I study in multiple solid tumors included a 53-patient cohort with NSCLC. In this subgroup of patients, the ORR was 21%. In this patient population, response to treatment with atezolizumab was found to be statistically significantly associated with tumor-infiltrating immune cell PD-L1 expression, but not tumor cell PD-L1 expression. The authors concluded that this suggested that PD-L1–suppressed preexisting immunity must be present in order to achieve an optimal response to PD-L1 blockade. Across all patient cohorts, rates of treatment-related grade 3/4 AEs were observed in 13% of patients. The phase II POPLAR trial assessed efficacy and safety of atezolizumab vs docetaxel in previously treated advanced NSCLC, with patient stratification based on PD-L1 tumor-infiltrating immune cell status, tumor histology, and previous lines of therapy (n = 277). OS was significantly improved in the atezolizumab population compared with the docetaxel population (12.6 months vs 9.7 months), and increasing improvement in OS was associated with increasing PD-L1 expression in tumor cells and/or tumor-infiltrating immune cells. This expression was defined differently and used a different assay compared with that used in the KEYNOTE trials; in the POPLAR trial it was defined via immunohistochemical scoring by PD-L1–expressing tumor-infiltrating immune cells as the percentage of tumor area: IC0 < 1%, IC1 ≥ 1% and < 5%, IC2 ≥ 5% and < 10%, and IC3 ≥ 10%; and as the percentage of PD-L1–expressing tumor cells: TC0 < 1%, TC1 ≥ 1% and < 5%, TC2 ≥ 5% and < 50%, and TC3 ≥ 50%. Treatment-related grade 3/4 AEs were seen in 11% of the atezolizumab-treated patients and in 39% of the docetaxel-treated patients, with one treatment-related death (from cardiac failure) in the atezolizumab group.
Ipilimumab, a fully human IgG1 anti–CTLA-4, has shown impressive outcomes in melanoma, and in some cases, durable responses. However, this agent has been less well studied in lung cancer, due to the observance of relatively higher rates of AEs and a relatively similar ORR when compared with PD-1/PD-L1 blockade in this patient population. Therefore, at this time it is more commonly being evaluated in combination with other agents used in the treatment of NSCLC. A phase II study in advanced or metastatic treatment-naive NSCLC evaluated ipilimumab in combination with platinum-based doublet chemotherapy. The authors compared ipilimumab administered concurrently with chemotherapy (4 cycles of carboplatin, paclitaxel, and ipilimumab, followed by 2 cycles of chemotherapy only), phased ipilimumab (2 cycles of chemotherapy alone followed by 4 cycles of chemotherapy with ipilimumab), and a control arm (6 cycles of chemotherapy alone). The study met its primary endpoint of improved immune-related PFS for phased ipilimumab compared with the control (5.7 months vs 4.6 months, respectively), but not for concurrent ipilimumab compared with the control (5.5 months vs 4.6 months, respectively). Median PFS for the phased ipilimumab, concurrent ipilimumab, and control groups was 5.1 months, 4.1 months, and 4.2 months, respectively; median OS was 12.2 months, 9.7 months, and 8.3 months, respectively; and rates of grade 3/4 immune-related AEs were 15%, 20%, and 6%, respectively. One treatment-related death in the concurrent ipilimumab arm was secondary to septic shock in the setting of toxic epidermal necrolysis.
A recently reported phase Ib safety and efficacy study evaluated durvalumab, a fully human IgG1 anti–PD-L1 antibody, in combination with tremelimumab, a fully human IgG2 anti–CTLA-4 antibody, in patients with advanced or metastatic NSCLC. The maximum tolerated dose was reached and appeared to be dependent on the dose of tremelimumab; across all dose levels, the rate of treatment-related serious AEs was 36%, with three treatment-related deaths (myasthenia gravis, pericardial effusion, and neuromuscular disorder). In the cohort treated with the dosing regimen selected for further study (durvalumab at 10 to 20 mg/kg every 2 to 4 weeks plus tremelimumab at 1 mg/kg), the ORR was 23% and the disease control rate was 35%. Across all dosing regimens, clinical activity was noted irrespective of tumor PD-L1 status.