Ariadna Juarez-Garcia1, Ruchika Sharma2, Matthias Hunger3, Sheena Kayaniyil4, John R Penrod5, Christos Chouaïd6. 1. Bristol Myers Squibb, Uxbridge, UK. 2. ICON plc., Bangalore, India. 3. ICON plc., Munich, Germany. 4. ICON plc., Toronto, ON, Canada. 5. Bristol Myers Squibb, Princeton, NJ, USA. 6. Centre Hospitalier Intercommunal (CHI) Créteil, Créteil, France. Electronic address: christos.chouaid@chicreteil.fr.
Abstract
BACKGROUND: Clinical trials have shown immunotherapy (IO) to be more effective than chemotherapy in pre-treated, advanced non-small cell lung cancer (NSCLC). However, there is a lack of understanding of its effectiveness in clinical practice, and among patient groups that are often underrepresented in trials. We aimed to summarize the existing real-world evidence (RWE) on the survival outcomes of IO in second- or higher line in advanced NSCLC. METHODS: We conducted a systematic review of real-world observational studies that reported overall survival (OS) estimates with IO, primarily nivolumab, pembrolizumab or atezolizumab, in adult, previously treated advanced or recurrent NSCLC patients. Meta-analysis was conducted using random-effect models to pool 1- and 2-year OS rates across studies. Additional subgroups were examined among patients treated with IO, including the elderly, those with poor performance status (PS) and those exhibiting metastasis. RESULTS: In total, 66 studies were included, of which 46 (70%) included a nivolumab-specific study arm. Pooled 1-year and 2-year OS rates with nivolumab monotherapy were 45.6% (95% CI; 43.4-47.8) and 28.0% (95% CI; 24.8-31.4), respectively, compared to 43.9% (95% CI; 39.1-48.8) and 20.4% (95% CI; 14.7-27.6) in the mixed immune checkpoint inhibitors (ICI) group. OS rates with nivolumab were slightly lower in elderly compared to non-elderly populations. Poor PS was associated with worse survival rates, with a pooled one-year OS estimate of 27.1% in PS ≥ 2 vs 51.6% in PS < 2. The pooled 2-year OS rate with nivolumab in patients with and without brain metastases was 22.1% and 26.1% respectively, and this difference was significant in 36% of individual studies. CONCLUSIONS: While the OS benefits of IO seen in real-world studies among pre-treated, advanced NSCLC patients are consistent with pivotal clinical trials, these tend to vary for the more vulnerable patient groups, such as patients with poor PS, which are often excluded from trials. Further research is needed to investigate findings in patients with brain and liver metastases.
BACKGROUND: Clinical trials have shown immunotherapy (IO) to be more effective than chemotherapy in pre-treated, advanced non-small cell lung cancer (NSCLC). However, there is a lack of understanding of its effectiveness in clinical practice, and among patient groups that are often underrepresented in trials. We aimed to summarize the existing real-world evidence (RWE) on the survival outcomes of IO in second- or higher line in advanced NSCLC. METHODS: We conducted a systematic review of real-world observational studies that reported overall survival (OS) estimates with IO, primarily nivolumab, pembrolizumab or atezolizumab, in adult, previously treated advanced or recurrent NSCLC patients. Meta-analysis was conducted using random-effect models to pool 1- and 2-year OS rates across studies. Additional subgroups were examined among patients treated with IO, including the elderly, those with poor performance status (PS) and those exhibiting metastasis. RESULTS: In total, 66 studies were included, of which 46 (70%) included a nivolumab-specific study arm. Pooled 1-year and 2-year OS rates with nivolumab monotherapy were 45.6% (95% CI; 43.4-47.8) and 28.0% (95% CI; 24.8-31.4), respectively, compared to 43.9% (95% CI; 39.1-48.8) and 20.4% (95% CI; 14.7-27.6) in the mixed immune checkpoint inhibitors (ICI) group. OS rates with nivolumab were slightly lower in elderly compared to non-elderly populations. Poor PS was associated with worse survival rates, with a pooled one-year OS estimate of 27.1% in PS ≥ 2 vs 51.6% in PS < 2. The pooled 2-year OS rate with nivolumab in patients with and without brain metastases was 22.1% and 26.1% respectively, and this difference was significant in 36% of individual studies. CONCLUSIONS: While the OS benefits of IO seen in real-world studies among pre-treated, advanced NSCLC patients are consistent with pivotal clinical trials, these tend to vary for the more vulnerable patient groups, such as patients with poor PS, which are often excluded from trials. Further research is needed to investigate findings in patients with brain and liver metastases.