A Review of Response Pathways and Workflow Integration Scope in Digital Intelligence Platforms for Breast Cancer Symptom Management

Breast cancer is the most common malignancy among women globally. Its treatment process is prolonged and complex, often accompanied by a range of physiological and psychological symptoms such as pain, fatigue, nausea, anxiety, and depression. These symptoms severely impact patients' quality of life, treatment adherence, and even survival outcomes. Traditional symptom management models predominantly rely on patients' voluntary reporting during follow-up visits, which is plagued by issues such as information delay, reporting bias, and delayed intervention. Consequently, many mild-to-moderate symptoms fail to be addressed in a timely manner, potentially escalating into severe adverse events requiring urgent intervention. Research by He et al. (2023) [1] also reveals the heavy symptom burden and significant unmet management needs faced by breast cancer patients during chemotherapy intervals. Therefore, establishing an active, continuous, and efficient symptom management system is a critical component of optimizing holistic breast cancer care. In recent years, the patient-centered philosophy has gained widespread acceptance. Patient-reported outcomes (PROs), defined as direct evaluations of health status from patients without clinical interpretation by healthcare providers, have increasingly demonstrated their value in both clinical practice and research. Reviews by Liu et al. (2013) [2] and Shi et al. (2023) [3] indicate that PROs can more sensitively capture the occurrence and fluctuation of symptoms compared to clinician reports, providing crucial evidence for clinical decision-making. With the proliferation of mobile internet, big data, and artificial intelligence technologies, digital health management platforms based on electronic PROs (ePROs) have emerged. Utilizing tools such as smartphone applications (Apps), web portals, or WeChat mini-programs, these platforms enable patients to report symptoms conveniently and in real-time, thereby breaking down the information barriers between in-hospital and out-of-hospital settings, as well as between treatment and rehabilitation phases. Systematic reviews by Aapro et al. (2020) [4] and Kirsch et al. (2024) [5] both confirm the significant potential of digital health tools in oncology supportive care, demonstrating their ability to improve symptom control, reduce emergency department visits and hospitalizations, and enhance quality of life. However, despite the proliferation of digital-intelligent symptom management platforms, there exist substantial variations in their design concepts, functional modules, response mechanisms, and workflow integration approaches. These variations have led to marked discrepancies in their clinical efficacy and scalability for widespread implementation. An effective digital-intelligent platform should not only achieve accurate symptom monitoring but also establish well-defined response pathways and efficient closed-loop intervention systems. This entails addressing a series of core questions: After the system receives a patient-reported outcome, who will be responsible for the response (physicians, nurses, artificial intelligence (AI), or the patients themselves)? What modalities will be adopted for the response (telephone calls, instant messages, automatic delivery of educational materials, or treatment adjustments)? How can the response process be seamlessly integrated with the hospital’s existing workflows (e.g., electronic health record (EHR) systems)? What are the actual application effects of these platforms, and what bottlenecks are encountered in their popularization and promotion? To address these questions, this review aims to systematically synthesize the existing literature evidence regarding the functional design, response pathways, and scope of workflow integration of digital symptom management platforms for breast cancer. This paper will propose a closed-loop classification framework centered on "Monitoring – Stratification – Response – Intervention," categorizing existing platforms into four main types and comparing their characteristics, supporting evidence, and limitations. Finally, this review will summarize the key challenges currently facing the field and provide an outlook on future development directions. The goal is to offer theoretical insights and practical guidance for developing and implementing more efficient and sustainable digital symptom management solutions for breast cancer in our national context.