Oral Presentation 2025 Joint Meeting of the COSA ASM and IPOS Congress

Symptom clusters in 40,000 Danish survivors of cancer: a cross-sectional study using exploratory factor analysis in population-based data (126123)

Anne Katrine Levinsen 1 , Anders Tolver 1 , Trille Kjaer 1 , Lau Caspar Thygesen 2 , Jan Wohlfahrt 1 , Susanne Rosthøj 1 , Lærke Kjær Tolstrup 3 , Marianne Nord Hansen 4 , Simone Oerlemans 5 , Christine Miakowski 6 , Peer Christiansen 7 , Peter Christensen 8 , Christoffer Johansen 9 , Robert Zachariae 7 , Lena Saltbæk 10 , Susanne Dalton 1
  1. Danish Cancer Institute, Copenhagen, Denmark
  2. National Institute of Public Health, Copenhagen
  3. Late effects Outpatient Clinic, Department of Oncology, Odense
  4. Late Effect Organization, Copenhagen
  5. Netherlands Comprehensive Cancer Organisation , Eindhoven
  6. University of California, School of Nursing & School of Medicine, San Francisco
  7. Danish Breast Cancer Group Center and Clinic for Late Effects, Århus
  8. Danish Cancer Society Centre for Research on Survivorship and Late Adverse Effects After Cancer in the Pelvic Organs, Århus
  9. Cancer late effects, Rigshospitalet, Copenhagen
  10. Danish Research Center for Equality in Cancer, Department of Clinical Oncology & Palliative Care, Næstved

Background: Late effects after cancer treatment challenge the well-being of cancer survivors. However, previous late effects research has focused on single symptoms. This approach may not capture the complex nature of late effects. This study aims to identify symptom clusters based on self-reported symptoms in 2-12-year cancer survivors.

Methods: The study is based on the nationwide SEQUEL questionnaire study and includes 40,766 survivors of cancer (i.e., breast, prostate, colorectal, lung, melanoma, lymphoma, and head and neck (HNC)) diagnosed between 2010-2019. Twenty-two symptoms commonly associated with cancer and its treatment were assessed using validated instruments (EORTC QLQ-C30, EORTC Item Library, GAD-7, PHQ-9). Symptoms severity was dichotomized into none/low versus moderate/severe. Exploratory factor analysis (EFA) was used to inform a structural equation model (SEM) for concurrent symptoms. Model predictions on test data were used to assess differences in symptom burden within specific symptom clusters across cancer sites. Analyses are ongoing.

Results: Dyspnea (31%), pain (29%), diarrhea (22%), and fatigue (21%) were the most prevalent symptoms across cancer sites. Given the presence of a specific symptom, the average number of co-occurring symptoms varied from 4 to 9. Preliminary results identified six clusters across cancer sites: pain-fatigue cluster, chemotherapy-related symptoms cluster, gastrointestinal symptoms cluster, psychological symptoms cluster, symptoms in the head and neck region cluster, and respiratory symptoms cluster. Findings from the SEM showed a difference in symptom burden across the six clusters by cancer site. Survivors of lung cancer, HNC, and lymphoma had a systematically elevated symptom burden for all symptom clusters compared to other cancer sites.  

Conclusions: Preliminary findings suggest that symptoms cluster together and that survivors from all included cancer sites report a high symptom burden within the identified clusters. These results support the need for identification of high-risk groups and monitoring of both single and multiple symptoms in follow-up care.