Cancer treatment: understanding risks and side effects

Towards shorter hospital stays: researchers explore options for outpatient stem cell therapy

Multiple myeloma is a cancer in which plasma cells, which normally produce antibodies, multiply uncontrollably in the bone marrow. There is currently no cure. However, various therapies can stabilise the disease and alleviate symptoms. One such therapy is to treat the patient with their own stem cells. This often involves weeks in hospital. Using machine learning methods, a research team has now assessed the conditions in which some of the therapy can be safely carried out as an outpatient. The study was conducted by researchers from the Göttingen Campus Institute for Dynamics of Biological Networks (CIDBN) at the University of Göttingen, the University Medical Center Göttingen (UMG), and the University Medical Center Bielefeld (OWL). It was published in the journal npj digital medicine.

The treatment, known as "autologous stem cell transplantation", involves taking stem cells from the blood of the patient. In preparation, the patient undergoes chemotherapy, and then a phase of stem cell mobilization in which stem cells are flushed from the bone marrow into the blood so that they can be collected. During the mobilization phase, patients often spend two to three weeks in hospital so that any serious side effects, such as kidney failure or infections, can be treated immediately. However, some patients develop serious side effects very late or not at all. "We asked ourselves whether the long hospital stay was really necessary for all patients," explains Dr Enver Aydilek, who works at UMG and OWL.

The researchers evaluated the treatment data of 109 patients with multiple myeloma who had undergone stem cell mobilization at the UMG. Using machine learning methods, they were able to identify time windows in which most patients were highly unlikely to suffer serious side effects. In the next step, they used the data to develop models to predict as accurately as possible when and which side effects might occur in which individuals.

For certain types of side effects, the models made highly accurate predictions, enabling a more precise risk assessment. First author Friedrich Schwarz, who studies medicine and data science at the UMG and Göttingen University and is a research associate at the CIDBN and the UMG Clinic for Haematology and Medical Oncology explains, "Our data-based therapy roadmap enabled us to better assess who will need inpatient monitoring and who can safely undergo chemotherapy and the mobilization phase whilst being closely monitored and cared for as an outpatient." This means the therapy is more patient-friendly and modern.

According to the research team, simulations of various treatment pathways show that outpatient treatment could be doubly beneficial. On the one hand, therapy can be tailored more to the individual and be less stressful, as patients’ quality of life is higher in their familiar home environment. At the same time, clinics can plan in a more resource-efficient manner. "Of course, the right conditions must be created for effective interaction between outpatient and inpatient care," emphasizes Schwarz. The study puts the first building blocks in place for this.

Original Publication: Schwarz, F., Levien, L., Maulhardt, M., Wulf, G., Brökers, N., & Aydilek, E. Predicting adverse events for risk stratification of chemotherapy based stem cell mobilization in multiple myeloma. npj digital medicine (2026). DOI: 10.1038/s41746-026-02394-y