126 - Biopsychosocial Predictors of Gastrointestinal Late Effects among Adolescent and Young Adult Survivors of Childhood Cancer

Publication Number: 2121.126

Nikhil Grobes, Childrens Hospital of Philadelphia, Philadelphia, PA, United States; Sara King-Dowling, Children's Hospital of Philadelphia, Philadelphia, PA, United States; Sarah N. Drake, Nationwide Children's Hospital, Columbus, OH, United States; Kelsey Woodard, Childrens Hospital of Philadelphia, Milford, NJ, United States; Dava Szalda, Childrens Hospital of Philadelphia, Philadelphia, PA, United States; Wendy Hobbie, The Children's Hospital of Philadelphia, Wynnewood, PA, United States; Jill P. Ginsberg, Children's Hospital of Pennsylvania/UPENN, Philadelphia, PA, United States; David Freyer, Children's Hospital Los Angeles, Los Angeles, CA, United States; Ahna L. Pai, Nationwide Children's Hospital, Columbus, OH, United States; Lisa Schwartz, Children's Hospital of Philadelphia/UPENN, Philadelphia, PA, United States

Background: Survivors of childhood cancer are at risk for gastrointestinal (GI) late effects of treatment (LE), which can span from the abdominopelvic cavity to the upper chest cavity (e.g., swallowing, digestion, bowel problems). They are understudied among adolescent and young adult (AYA) survivors.

Objective: To describe the prevalence of and biopsychosocial associates of GI LE among AYA survivors.

Design/Methods: AYA (N=587) 16-25 y/o, ≥ 2 years off treatment, and ≥ 5 years since diagnosis [M(SD) age = 19.7(2.5); 52.5% female, 61.8% Non-Hispanic White] participated in a longitudinal study of self-management. GI LE were abstracted from the electronic health record (EHR) from the last oncology note and self-reported via the Health Knowledge Inventory, which asks survivors to check problems related to their cancer history from a list of LE. Potential demographic, disease/treatment, and psychosocial predictors were extracted from the EHR or patient-reported. We conducted univariate (chi-square and t-tests) analyses, followed by logistic regressions, entering significant univariate correlates.

Results: AYA (n=100; 17%) self-reported having GI LE; 36 (6%) had GI LE documented in the EHR; twenty (3.4%) had both self-reported and EHR-confirmed GI LT. Associates (p <.05) of EHR-reported GI LE included any radiation, chest or abdominal radiation, transplant, and solid tumor, with a trend for neighborhood income. Self-reported GI LE were associated with being female, older age, total body radiation, transplant, anxiety, depression, and cancer worry. Logistic regressions revealed that EHR-reported GI LE were more likely among patients from low-income neighborhoods (Exp(B)=2.39, p=.03) and who received radiation (Exp(B)=4.12, p<.001). Self-reported GI LE were more likely among AYA who received total body radiation (Exp(B)=2.62, p=.005), or with greater cancer worry (Exp(B)=1.02, p<.001).

Conclusion(s): Self-reported GI LE were more prevalent than reported in the EHR, indicating that many AYA may be inadvertently misattributing GI symptoms to cancer history. As expected, radiation was significantly related to GI LE as it is a known risk factor. Anxiety and cancer worry were related to self-reported GI LE. This finding supports the well-established relationship between stress and GI function. Neighborhood income was related to EHR-documented LE which is consistent with research on general GI symptoms, but is puzzling that this was not an associate of self-reported GI LE. More research is needed to understand what is driving GI symptoms among AYA survivors and to target modifiable variables (e.g., cancer worry) to mitigate them.