PEOPLE (NTC03447678), a phase II trial to test pembrolizumab as first-line treatment in patients with advanced NSCLC with PD-L1 <50%: a network analysis

Background In the dynamic landscape of cancer treatment, immunotherapy plays a pivotal role, particularly for advanced non-small cell lung cancer (aNSCLC) lacking driver alterations. While PD-L1 Tumor Proportion Score (TPS) guides patient selection, challenges persist in refining strategies. The phase II PEOPLE trial addresses this by exploring biomarkers in aNSCLC patients with PD-L1 TPS &lt; 50% treated with pembrolizumab. Leveraging network medicine, which maps complex biological interactions, this study integrates high&#x2;throughput technologies to analyze circulating immuno-profiling and gene expression data. The aim is to unveil novel biomarkers critical for immunotherapy response, fostering personalized treatments. This interdisciplinary collaboration between clinicians and data scientists is essential for decoding aNSCLC complexities. The thesis outlines objectives: constructing correlation networks and identifying predictive biomarkers. Methods: To explore this challenge, network analysis techniques were employed. The work used data from the phase II trial PEOPLE (NCT03447678) conducted at the IRCCS Fondazione Istituto dei Tumori di Milano. This dataset contained comprehensive information on patients’ circulating immune profiles and gene expression profiles. Data preprocessing steps ensured quality, and statistical analyses involved network construction, differential expression analysis, and enrichment analysis. Additionally, community detection was applied to the differential co-expression network. Patient similarity networks were created, and survival analysis was conducted to understand the impact of identified biomarkers on overall survival (OS). Results: Survival analysis, with a median follow-up of 26.4 months, revealed a median PFS of 2.9 months and a median OS of 12.1 months. Response rates were 24.1%, and disease control rate was 53.4%. Differential correlation networks (DCN) of circulating immune profiling demonstrated distinct patterns in responders and non-responders pre-therapy, shedding light on key role of NK cells. Gene DCN identified 23 hubs gene and enrichment analysis of each hub revealed associations with immune-related processes. Community detection identified modules enriched in immune responses. A patient similarity network based on hub genes revealed two clusters with significant differences in survival outcomes, emphasizing the potential prognostic value of molecular heterogeneity in response to pembrolizumab. Conclusions: These findings contribute valuable insights to the evolving landscape of immunotherapy in aNSCLC, emphasizing the need for further investigations into the intricate relationships shaping treatment response and patient outcomes. Background In the dynamic landscape of cancer treatment, immunotherapy plays a pivotal role, particularly for advanced non-small cell lung cancer (aNSCLC) lacking driver alterations. While PD-L1 Tumor Proportion Score (TPS) guides patient selection, challenges persist in refining strategies. The phase II PEOPLE trial addresses this by exploring biomarkers in aNSCLC patients with PD-L1 TPS &lt; 50% treated with pembrolizumab. Leveraging network medicine, which maps complex biological interactions, this study integrates high-throughput technologies to analyze circulating immuno-profiling and gene expression data. The aim is to unveil novel biomarkers critical for immunotherapy response, fostering personalized treatments. This interdisciplinary collaboration between clinicians and data scientists is essential for decoding aNSCLC complexities. The thesis outlines objectives: constructing correlation networks and identifying predictive biomarkers. Methods: To explore this challenge, network analysis techniques were employed. The work used data from the phase II trial PEOPLE (NCT03447678) conducted at the IRCCS Fondazione Istituto dei Tumori di Milano. This dataset contained comprehensive information on patients’ circulating immune profiles and gene expression profiles. Data preprocessing steps ensured quality, and statistical analyses involved network construction, differential expression analysis, and enrichment analysis. Additionally, community detection was applied to the differential co-expression network. Patient similarity networks were created, and survival analysis was conducted to understand the impact of identified biomarkers on overall survival (OS). Results: Survival analysis, with a median follow-up of 26.4 months, revealed a median PFS of 2.9 months and a median OS of 12.1 months. Response rates were 24.1%, and disease control rate was 53.4%. Differential correlation networks (DCN) of circulating immune profiling demonstrated distinct patterns in responders and non-responders pre-therapy, shedding light on key role of NK cells. Gene DCN identified 23 hubs gene and enrichment analysis of each hub revealed associations with immune-related processes. Community detection identified modules enriched in immune responses. A patient similarity network based on hub genes revealed two clusters with significant differences in survival outcomes, emphasizing the potential prognostic value of molecular heterogeneity in response to pembrolizumab. Conclusions: These findings contribute valuable insights to the evolving landscape of immunotherapy in aNSCLC, emphasizing the need for further investigations into the intricate relationships shaping treatment response and patient outcomes.