Comparative investigation of neoadjuvant immunotherapy versus adjuvant immunotherapy in perioperative patients with cancer: A metrology informatics analysis based on machine learning

BackgroundFor cancer neoadjuvant and adjuvant immunotherapy, after several decades of evolution, the field currently possesses an enormous volume of underutilized data. Informatics analysis to thoroughly excavate the similarities and differences between the two is desperately necessary.MethodsExtensive relevant studies (n=1373) on neoadjuvant and adjuvant immunotherapy from 2014-2023 were collected for quantitative, hierarchical clustering, and comparative analyses after vigorous quality control.ResultsOver the last decade, neoadjuvant and adjuvant immunotherapy enjoyed promising development status (Annual Growth Rate: 25.18% vs 6.52%) and global collaboration (International Co-authorships: 19.93% vs 19.84%). Unsupervised hierarchical clustering identified their dominant research clusters, in which Cluster 4: Balance of neoadjuvant immunotherapy efficacy and safety and Cluster 2: Adjuvant immunotherapy clinical trials are emerging research populations. Burst and regression curve analyses uncovered domain pivotal research signatures, including biomarkers (R2=0.6505, p=0.0086) in neoadjuvant scenarios, and tumor microenvironment (R2=0.5571, p=0.0209) in adjuvant scenarios. The Walktrap algorithm further revealed that "non-small cell lung cancer, immune checkpoint inhibitors, melanoma" and "melanoma, hepatocellular carcinoma, dendritic cells" (Relevance Percentage: 100% vs 100%, Development Percentage: 37.5% vs 17.1%) are extensively relevant to this field, but remain underdeveloped. Furthermore, comprehensive quantitative comparisons revealed that this field's spotlight on neoadjuvant immunotherapy overtook adjuvant immunotherapy entirely after 2020; such a qualitative finding will facilitate proper decision-making for subsequent research and avoid significant wastage of healthcare resources.ConclusionsThis cross-sectional study comparatively analyzed the fundamental metrological information in cancer neoadjuvant and adjuvant immunotherapy, identified their pivotal research signatures, and provided some substantial predictions for their subsequent preclinical and clinical research.Legal entity responsible for the studyThe authors.FundingHas not received any funding.DisclosureAll authors have declared no conflicts of interest.