Basic science discoveries and cancer treatment innovations have led to 15.5 million cancer survivors
currently in the United States, and the number of survivors is estimated to reach 20 million by 2026. Cancer and
its treatment often cause long-term effects, including both persistent symptoms and functional deficits that
negatively impact survivors’ quality of life. Cancer survivors are an ideal population in which to examine multiple
co-occurring symptoms, termed “symptom clusters.” A 2017 panel of experts in symptom science convened by
the National Institute of Nursing Research (NINR) and the Office of Rare Diseases noted that patients with
chronic conditions such as cancer experience an array of symptom clusters that have a negative impact on
health-related quality of life (HRQOL). This panel called for advancing symptom science through symptom cluster
research to build the evidence base for effective symptom management. Our proposal is in response to an
NINR/NCI joint program announcement that emerged from this workshop. Most prior research on symptom
clusters in cancer has focused on patients undergoing the acute phase of cancer treatment for specific common
cancers often treated in academic referral centers. Our study will address these limitations by focusing on early
stage, sociodemographically diverse adult cancer survivors treated in general US oncology practices.
We will investigate symptom clusters using data from a previously surveyed cohort of 5,506 adult cancer
survivors initially diagnosed between 21-84 years of age with 1 of 7 distinct cancer types who were recruited in
partnership with 4 population-based Surveillance, Epidemiology, and Ends Results (SEER) cancer registries in 3
states (CA, NJ, LA). This cohort was first surveyed at a median of 9 months after initial cancer diagnosis with a
follow-up survey conducted 6 months later. The SEER registry data include detailed clinical and treatment data,
and the survey collected information on 8 domains of symptoms and functional status common and highly
impactful in cancer survivors, and numerous other sociodemographic and economic variables. In Aim 1, we will
identify the prevalence of symptom clusters and subgroup profiles using innovative psychometric and statistical
methods. We will evaluate sociodemographic (e.g., age, sex, race-ethnicity, income, education) and clinical
characteristics (e.g., cancer type, stage, treatment, and comorbidity) associated with each symptom cluster and
survivor subgroup. In Aim 2 we will investigate the temporal patterns of symptom clusters using state-of-the-art
latent transition analysis. In both aims, we will evaluate the role of several mutable factors at the patient and
healthcare system levels to inform future interventional research. In Aim 3 we will investigate the association of
evidence-based symptom and comorbidity care with the prevalence and trajectory of symptom clusters using
linked Medicare claims data for the subgroup of survivors ages 65 and over. Our results will enable a richer
understanding of the phenomenon of symptom clusters and will provide a new foundation of knowledge that will
help inform future interventions to mitigate the adverse effects of symptom clusters.