Benefit-Risk Summary of Nivolumab for Patients With Metastatic Squamous Cell Lung Cancer After Platinum-Based Chemotherapy: A Report From the US Food and Drug Administration

Importance  Metastatic squamous non–small-cell lung cancer (SQ NSCLC) is a serious and life-threatening malignant condition with unmet medical need. In late December 2014, the US Food and Drug Administration (FDA) obtained the data monitoring committee report of a planned interim analysis of a trial in second-line SQ NSCLC (CM017) that demonstrated an overall survival benefit for patients treated with nivolumab compared with docetaxel.

Observations  In that trial, 272 patients with metastatic SQ NSCLC patients had been randomized to receive nivolumab (n = 135) or docetaxel (n = 137). Median overall survival was 9.2 months for patients randomized to nivolumab and 6.0 months for those randomized to docetaxel (hazard ratio, 0.59; 95% CI, 0.44-0.79; P < .001). The safety of nivolumab was evaluated in a single-arm trial of 117 patients in previously treated metastatic SQ NSCLC and was consistent with the safety profile in melanoma, with rare but serious immune-mediated adverse events managed with corticosteroids and dose interruption.

Conclusions and Relevance  The FDA granted nivolumab traditional approval on March 4, 2015, for treatment of metastatic SQ NSCLC with progression during or after platinum-based chemotherapy. The approval provides an important treatment option for these patients, affecting routine care and clinical trials.

Lung cancer is the leading cause of worldwide cancer-related deaths and is usually diagnosed at advanced stages when the prognosis is poor.¹ In contrast to the substantial progress seen in the development of targeted therapies against tumors with oncogene addiction (eg, EGFR and ALK alterations, most often found in adenocarcinoma non–small-cell lung cancer [NSCLC]), there has been little progress in the development of highly effective treatments targeting genetic alterations predominantly found in metastatic squamous cell (SQ) NSCLC.^2-8Table 1 highlights drugs approved by the US Food and Drug Administration (FDA) for second-line SQ NSCLC.

New drugs focus on targeting the immune system.¹² Pathways involved in inhibiting antitumor T-cell responses (activation of the inhibitory coreceptors cytotoxic T-lymphocyte–associated protein 4 and programmed cell death protein 1 [PD-1] on T cells) allow tumors to evade the immune system. Nivolumab (Opdivo; Bristol-Myers Squibb Company), a monoclonal antibody directed against PD-1, is one of the first of a class of drugs that block T-cell inhibitory signal pathways by preventing engagement of PD-1 to its ligands. Recent high-throughput genomic analyses have shown that NSCLC has a substantial amount of DNA damage, molecular heterogeneity, and mutational burden (reflecting tobacco carcinogen exposure). It is thought that the increase in mutational burden gives rise to neoantigens and tumor immunogenicity, which is important for tumor sensitivity to PD-1 axis blockade.^13-15

Box Section Ref ID

Two studies of nivolumab assessing efficacy in metastatic SQ NSCLC have been reported. Study CA209063 (CM063) was a single-arm, multinational trial in patients with metastatic SQ NSCLC (n = 117), which had progressed after treatment with 2 systemic regimens including platinum-based doublet chemotherapy and is well described by Rizvi et al.¹⁶ Patients received nivolumab, 3 mg/kg, as an intravenous infusion every 2 weeks until progression or occurrence of toxic effects. The primary outcome was Response Evaluation Criteria in Solid Tumors version 1.1¹⁷ objective response rate (ORR), as determined by a blinded independent review committee.

In April 2014, the FDA determined that based on the “modest” ORR (15%) in CM063, further information from the ongoing randomized study should be provided to support a regulatory decision. The FDA requested the results of a planned interim analysis of the primary end point, overall survival (OS), from the randomized trial (study CA209017 [CM017]), well described by Brahmer and colleagues.¹⁸ On December 19, 2014, the FDA received the external data monitoring committee (DMC) report, which stated that the O’Brien-Fleming boundary¹⁹ had been crossed and that nivolumab demonstrated superior OS vs docetaxel. Based on these results, the FDA informed Bristol-Myers Squibb to submit the final component of the Biologics License Application (BLA) for nivolumab. On January 10, 2015, the DMC statisticians presented the report to the full DMC, who recommended that, in light of the positive nivolumab results, CM017 patients randomized to the docetaxel arm be allowed to cross over to receive nivolumab.

The CM017 study was an open-label, multicenter, multinational, randomized (1:1) trial in patients whose disease had progressed during or after 1 prior platinum-based chemotherapy regimen.¹⁸ Key eligibility criteria included histologically or cytologically documented SQ NSCLC with stage IIIB-IV disease (International Association for the Study of Lung Cancer, version 7²⁰) or with recurrent or progressive disease following multimodal therapy. Patients received nivolumab (n = 135), 3 mg/kg, intravenously every 2 weeks, or docetaxel (n = 137), 75 mg/m², intravenously every 3 weeks. The major efficacy outcome was OS.

Based on the efficacy results of the CM017 study, in which a statistically significant and clinically meaningful improvement in survival was demonstrated for nivolumab compared with docetaxel, the FDA committed to an expeditious review (<4 months) of the BLA compared with both the standard (10-month) and priority-designated (6-month) BLA review times.

In CM017, the prespecified interim analysis for the primary end point occurred when 199 deaths were observed (86% of the planned number of events for final analysis).¹⁸ The trial demonstrated a 41% decreased risk of death for the nivolumab arm. The median OS was 9.2 months for patients assigned to nivolumab and 6.0 months for those assigned to docetaxel (hazard ratio, 0.59; 95% CI, 0.44-0.79; P < .001). Specific survival data are summarized in Table 2, and the Figure shows how the Kaplan-Meier survival curves continue to separate beyond the median.

These results were supported by demonstration of antitumor activity in the single-arm trial CM063, in which the ORR for nivolumab was 14.5% (95% CI, 8.7%-22.2%).¹⁶ All responses were partial responses. At the time of analysis, 10 (59%) of 17 responding patients had durable responses of 6 months or longer, and the duration of response ranged from at least 1.9 to at least 11.5 months. In CM017, the reported ORR was 20% for nivolumab and 9% for docetaxel.¹⁸

To expedite the submission of the BLA, the FDA did not wait to receive or review safety data from CM017. However, the approval was contingent on a postmarketing requirement for formal submission of the CM017 data within the year. Therefore, the FDA relied primarily on safety data in patients with SQ NSCLC from the single-arm CM063 trial, supported by safety data obtained in other disease settings. The most common (≥30%) adverse reactions and laboratory abnormalities in the 117 patients receiving nivolumab in CM063 were fatigue, lymphopenia, dyspnea, decreased appetite, and cough.¹⁶ The most frequent (greater than or equal to 5%) grades 3 and 4 adverse reactions and laboratory abnormalities were dyspnea, fatigue, lymphopenia, and hyponatremia. Immune-mediated adverse reactions in CM063, defined as cases requiring use of systemic corticosteroids with no clear alternative cause were immune-mediated pneumonitis (6.0%), hypothyroidism (4.3%), hyperthyroidism (1.7%), motor dysfunction (1.7%), rash (1.7%), adrenal insufficiency (0.9%), vasculitis (0.9%), colitis (0.9%), and renal dysfunction (0.9%). Immune-mediated adverse reactions were managed with administration of high-dose (2- to 4-mg/kg prednisolone equivalent) corticosteroids followed by a taper and interruption of nivolumab therapy. No patients administered corticosteroids were rechallenged with nivolumab following corticosteroid taper. The FDA has mandated a postmarketing requirement to submit the results of the randomized CM017 trial to better characterize the incidence, severity, and outcomes of nivolumab-induced immune-mediated adverse reactions in patients with NSCLC.

A summary of the FDA’s benefit-risk assessment is presented in the Box. After observing the magnitude of the survival improvement over an active treatment, docetaxel, from the CM017 trial,¹⁸ the FDA believed that it was critical to incorporate the data from CM017 into the BLA and include these findings into the product label to support an indication for second-line treatment of SQ NSCLC, a disease with unmet need. Rapid dissemination of these data would ensure timely access to nivolumab for patients and physicians as well as a reevaluation of control arms for ongoing and planned trials in SQ NSCLC.²¹

Box Section Ref ID

The information from the CM017 trial submitted to the FDA was the survival and demographic data analyzed by the DMC under a prespecified interim analysis. The safety data from the single-arm CM063 trial, which in combination with the safety data from the prior approval in melanoma, provided sufficient basis to characterize risks to inform prescribers of safe use and confirm a favorable benefit-risk analysis, considering the magnitude of OS effect, to support approval. A conventional submission that included all the data (including safety) from the CM017 trial would have delayed the BLA submission. On average, sponsors require 6 months or more to clean, prepare, and standardize data sets, perform quality control and assurance, write clinical study reports, and submit elements from other disciplines, including chemistry, manufacturing and controls, nonclinical toxicology, and clinical pharmacology modules. As a condition of the approval, Bristol-Myers Squibb must submit a more comprehensive report and data from the CM017 trial.

This application was noteworthy for a number of reasons. It is the first head-to-head trial against an active control in second-line squamous NSCLC to demonstrate a large survival benefit and is the first immunotherapy approved for treatment of a histologic subtype of NSCLC. Survival is the gold standard for clinical benefit because it is objective and not influenced by potential bias that may occur with progression-free survival. The 15% ORR in the CM063 single arm study,¹⁶ though durable, was modest. Absent the data from the CM017 randomized trial, it is uncertain that the ORR data alone would have supported an accelerated approval, which would have required substantial discussion including advice obtained during an advisory committee meeting. Importantly, in CM017, the docetaxel control arm performed similarly to other recent studies in terms of OS; for example, in REVEL,¹¹ the median OS was 9.1 months (Table 1). It is unclear if ORR and duration of response fully capture the clinical benefit of immunotherapy; further research is needed to explore novel end points for activity estimation and assessment of clinical benefit.²²

In both the CM017 and CM063 studies, patients were entered irrespective of tumor PD-1 ligand 1 (PD-L1) status,^16,18 and further trials with PD-1/PD-L1 inhibitors will need to address the value of predictive biomarkers in NSCLC. In addition, precompetitive collaboration and cross-validation is needed for the various PD-L1 assays in clinical development.²³

The FDA’s approach to reviewing and incorporating the OS results from the CM017 study into product labeling ensured rapid access to patients with SQ NSCLC that progressed during or after platinum-based chemotherapy. Specifically, 6 months was saved by not waiting for formal preparation of the data by the sponsor, and an additional 2.5 months was saved by the FDA expedited review. Studies are ongoing to evaluate the role of PD-1/PD-L1 inhibitors in other NSCLC histologic types and disease settings.

Accepted for Publication: August 18, 2015.

Corresponding Author: Dickran Kazandjian, MD, Office of Hematology Oncology Products, US Food and Drug Administration, 10903 New Hampshire Ave NE, WO22 2320, Silver Spring, MD 20993 (Dickran.kazandjian@fda.hhs.gov).

Published Online: October 15, 2015. doi:10.1001/jamaoncol.2015.3934.

Author Contributions: Drs Kazandjian, Khozin, and Blumenthal contributed equally to this work. Dr Kazandjian had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Kazandjian, Blumenthal, Kluetz, Pazdur.

Acquisition, analysis, or interpretation of data: Kazandjian, Khozin, Zhang, Tang, Libeg, Sridhara, Keegan, Pazdur.

Drafting of the manuscript: Kazandjian, Khozin, Blumenthal, Zhang, Tang, Libeg, Pazdur.

Critical revision of the manuscript for important intellectual content: Kazandjian, Khozin, Kluetz, Sridhara, Keegan.

Statistical analysis: Zhang, Tang, Sridhara.

Administrative, technical, or material support: Kazandjian, Khozin, Libeg, Pazdur.

Study supervision: Kazandjian, Blumenthal, Kluetz, Sridhara, Keegan, Pazdur.

Conflict of Interest Disclosures: None reported.