AI-Driven Marine Bioactives for Cancer Therapy: A Systematic Review of Drug Discovery, Resistance Overcoming Strategies, and Precision Drug Delivery

Cancer continues to pose a significant global health challenge, with existing treatment options encountering obstacles such as drug resistance, systemic toxicity, and limited bioavailability. Marine ecosystems present a largely underutilized source of bioactive compounds that possess strong anticancer properties. The advent of Artificial Intelligence (AI) is transforming the landscape of drug discovery by improving the identification, optimization, and delivery processes for these marine-derived bioactives. This systematic review seeks to: (1) examine marine-derived anticancer bioactives along with their mechanisms of action and clinical significance; (2) evaluate AI-driven approaches in the discovery and screening of marine compounds; (3) investigate AI-based strategies for addressing drug resistance; and (4) assess AI-guided precision delivery methods for marine-derived anticancer agents. A thorough literature search was performed utilizing databases such as PubMed, Web of Science, Scopus, and Google Scholar. Studies were selected based on their investigation into marine bioactives with anticancer effects, AI-assisted drug discovery techniques, AI-enhanced mechanisms related to drug resistance, or AI-supported delivery systems. Data extraction centered on the sources of bioactives, molecular mechanisms involved, types of AI models applied, and stages of clinical translation. Notable marine-derived compounds like Trabectedin, Salinosporamide A, and Fucoidan demonstrate significant anticancer effects through mechanisms including apoptosis induction, angiogenesis inhibition, and targeting proteasomes. Computational models utilizing AI enhance the screening process for marine bioactives via high-throughput virtual screening methodologies, molecular docking analyses, and Quantitative Structure-Activity Relationship (QSAR) modeling. Analysis of drug resistance driven by AI identifies relevant biomarkers, refines therapeutic regimens, and forecasts tumor responses. Moreover, nanoparticle-based delivery systems optimized by AIalongside liposomes and hydrogel formulations-enhance drug stability, improve bioavailability levels, and increase efficiency in targeting tumors. AI is reshaping the field of cancer drug discovery from marine sources by expediting compound identification processes while optimizing treatment strategies and customizing drug delivery methods. Future investigations should prioritize expanding databases focused on AI-powered marine bioactives; refining predictive models applicable in clinical settings; and developing sustainable bioprospecting strategies driven by AI. Approaches guided by AI hold considerable promise in overcoming challenges associated with drug resistance while enhancing both precision and effectiveness in therapies derived from marine sources. A systematic review was conducted to assess the effects of AI-enhanced marine bioactives on cancer treatment. The study selection process is depicted in Figure 1 (PRISMA Flow Diagram). Initially, a total of 3,200 records were identified through searches in several databases, such as PubMed, Web of Science, Scopus, and Google Scholar. Additionally, 100 records were sourced from other avenues. After removing 600 duplicate entries, 2,700 studies remained for further screening. The titles and abstracts of these 2,700 records were carefully examined, leading to the exclusion of 2,000 records that did not meet the inclusion criteria. This included studies on irrelevant topics, those unrelated to cancer treatment, non-marine bioactives, or articles lacking AI methodologies. From the remaining articles, 700 full-text publications were assessed for eligibility. Following an in-depth evaluation process, 400 full-text articles were excluded due to various reasons such as inadequate AI integration, lack of relevance to marine bioactives, or being review articles instead of original research. In the end, a total of 300 studies were included in the qualitative synthesis (systematic review), while an additional 150 studies qualified for quantitative synthesis (meta-analysis where applicable). This PRISMA flow diagram (Figure 1) offers a clear and transparent depiction of the study selection methodology utilized in this research project, ensuring both reproducibility and methodological rigor.