Real‐world experience of nivolumab in the treatment of poor performance status patients with advanced non‐small cell lung cancer

Abstract Background Nivolumab improves disease control and survival in advanced NSCLC in patients with good performance status (PS), but there is limited data on its efficacy in patients with poor PS. Aim Primary objective of the study was to evaluate the efficacy and safety of nivolumab and examine the influence of PS on outcomes. Methods and Results Retrospective analysis of patients treated with single‐agent nivolumab for advanced NSCLC at a single institution was performed. Sixty‐six patients treated with nivolumab were identified (33 male) with a median age of 68.5 years. Fifty‐six (85%) patients were current or former smokers and 17 (26%) had brain metastasis. All patients had received prior chemotherapy, 39 (59%) patients received one and 27 (41%) had ≥2 prior lines of therapy. Median overall survival (OS) was 7.1 months (95%CI 3.61–11.3) in the overall study population. OS of patients with PS ≥2 at the start of treatment was 3.04 months (95%CI 1.64–7.36) as compared to 10.23 months (95%CI 7.06–18.9) with PS ≤1. The overall response rate was 7% (four patients had a partial response), 23 (40%) patients had stable disease; the overall disease control rate (partial response and stable disease) was 47%. Twenty‐six (40%) patients had PS ≥2 at the start of treatment and 2 (8%) of these patients developed a partial response, 4 (15%) had stable disease; the overall disease control rate was 23%. Fourteen (58%) patients with PS ≥2 had disease progression at the time of first disease response evaluation. In the overall population, 20% of patients experienced grade ≥3 treatment‐related adverse events (TRAEs), most commonly pneumonitis, hepatitis, and colitis. Fourteen TRAEs led to treatment discontinuation, 9 (23%) adverse events (AEs) in patients with PS ≤1 and 5 (19%) with PS ≥2. Fourteen (21%) patients died within 30 days of the last nivolumab treatment. Conclusion There was no significant difference in toxicity leading to treatment discontinuation between the poor and good PS groups, but survival was shorter with poorer PS. PS appears to be an important prognostic factor and remains a relevant discriminator in the selection of treatment with immunotherapy for lung cancer.

poorer PS. PS appears to be an important prognostic factor and remains a relevant discriminator in the selection of treatment with immunotherapy for lung cancer. OS rates of 17% with nivolumab as compared to 8% with docetaxel. 3 In Australia, nivolumab was initially available through a compassionate access program, and in 2016 it was approved by Pharmaceutical Benefits Scheme (PBS), the Australian national government drug funding program. Treatment efficacy of nivolumab in patients with advanced NSCLC has been demonstrated for patients with a good performance status (PS) (Eastern Cooperative Group (ECOG) ≤1). 4 The efficacy of immunotherapies in the setting of poor PS has not been extensively evaluated. Patients with a PS of two accounts for approximately 30%-40% of patients diagnosed with advanced NSCLC in clinical practice. 5 Patients with poor PS have historically been excluded or underrepresented in the clinical trials due to poor outcomes and there is a paucity of data to guide the optimal treatment strategy in this population.
Recently few prospective studies have explored the outcomes with the use of immunotherapy in advanced NSCLC patients with poor PS. Data from Phase 3B/4 Checkmate 153 study evaluating a subgroup of patients based on PS showed lower estimated 6 months OS in patients with PS 2 (41%) than PS 0-1 (65%), however, there was no significant difference in toxicities for poor PS patients compared to the overall study population. 6 Similarly, results from the Phase II Checkmate 171 study showed that patients with PS 2 had lower median OS, 3 months and 6 months survival rates as compared to the overall study population. 7 CHECKMATE 169 evaluated the advanced NSCLC patient outcomes that were treated with nivolumab and progressed after ≥1 prior line of prior therapy. This study included patients with age ≥ 70 years and PS 2 however patients with CNS metastasis were excluded. Preliminary results from the Canadian cohort of this study showed comparable safety outcomes and lower OS for patients with PS 2 as compared to the overall population. 8 We suspected that the use of nivolumab in the 'real-world' outside of the restrictive entry criteria of clinical trials, would include patients with poor PS, with both patients and oncologists enthusiastic about trying this novel treatment approach.
The objectives of the analysis were to explore the 'real-world' application of nivolumab with the evaluation of efficacy, safety with analysis of treatment-related toxicities, and effect of PS on outcomes particularly OS.  NSCLC and showed comparable outcomes as compared to the overall population in both analyses. In our study 28 (42%) patients were ≥ 70 years and the majority had PS ≤1 (n = 18) and 10 patients had PS ≥2. Seventeen patients (26%) who had CNS metastases out of these majority (n = 9) had PS ≥2. In the real-world experience management of elderly and frail patients as well as patients with CNS metastasis is challenging and although we did not evaluate the specific outcomes for these patients, our study population is indicative of these practical management issues. Further studies are required to specifically evaluate these subsets to help guide appropriate treatment with benefit.

| METHODS
One of the limitations of our study is the small population size from a single institution and due to this, we could not further evaluate the immunotherapy-related outcomes in a specific subset of patients in addition to the potential impact of the use of antibiotics and steroids on the outcomes. Despite these limitations, the heterogeneity of the real-world patient cohort and outcomes consistent with the existing data does add to the importance of considering PS as an important factor before starting the immunotherapy.
In conclusion, in this retrospective real word study, we found that patients with poor PS have shorter OS as compared to good PS, but there was no significant difference in tumor response rates or safety outcomes based on PS. Further studies are needed to define the optimal treatment approach for patients with poor PS.