Trajectories of multidimensional worry in survivors of early breast cancer (BC).

Abstract

12051 Background: Worrisome thoughts are common future-oriented behaviors and may affect quality of life (QoL) years removed from BC diagnosis. We aimed to identify trajectories of worry and their determinants, focusing on age-related specificities. Methods: Disease-free stage I-III BC survivors from CANTO (NCT01993498) completed the Impact of Cancer scale v2 at end of treatment (EoT) and years (Y) 1, 3, and 5 post-EoT. Group-based trajectories were identified modeling continuous scores of the worry subscale (range: 1-5, higher = worse), which evaluates uncertainty and worry about the future and health, a sense of time running out, fear of cancer recurrence (FCR), and symptom-triggered FCR. Adjusted multinomial logistic regression incorporating pre-treatment anxiety/depression assessed membership determinants. QoL (EORTC QLQ-C30/BR23) was described by trajectory. Analyses were stratified by age. Results: We included7824 survivors, 31% were ≤50 and 69% > 50 years old. Among those aged ≤50, we identified five worry trajectories: most had high (41%) or very high (18%) worry (mean scores [95% CI] at EoT: 3.57 [3.55-3.60] and 4.37 [4.33-4.41], respectively, stable); 21% showed improvement (2.77 [2.73-2.82] at EoT, 2.14 [2.08-2.21] at Y5); and 12% had low worry (1.58 [1.53-1.63] at EoT, stable). Finally, 7% had moderate worry at EoT (2.21 [2.14-2.28]) that worsened sharply, stabilizing at high levels by Y5 (3.40 [3.29-3.52]). These patterns, especially in the worsening group, were primarily driven by persisting FCR and increasing uncertainty about health. Among those aged > 50, similar trajectories emerged, however worry levels in the worsening group (17%) remained low-to-moderate (2.04 [2.01-2.06] at EoT, 2.34 [2.27-2.40] at Y5). Younger age was a significant determinant of very high worry, both in the ≤50 (Odds Ratio v low worry [95% CI] per 10-year decrease: 1.92 [1.33-2.77]) and in the > 50 age group (1.50 [1.20-1.90]). Partnered status was associated with worsening worry in the ≤50 age group (v not: 2.54 [1.24-5.19]), whereas chemotherapy with very high worry in the > 50 age group (v no chemo: 1.77 [1.14-2.73]). BC grade, stage, BRCA status, family history of cancer, and comorbidities were not linked to worry in either age group. Psychological and physical domains of QoL, including fatigue, sleep, cognitive function, appetite, and gastrointestinal symptoms, were significantly worse in trajectories at higher worry (p < .001 at all time points). Conclusions: More than two-thirds of younger BC survivors report high levels of worry that either remain unchanged or worsen 5 years post-EoT, with a more pronounced decline and different determinants than older counterparts. These patterns appear to be largely driven by the post-traumatic effects of BC on unresolved FCR and by negative projections about health. Early referral to mental health professionals and monitoring strategies focused on reducing anxious arousal are warranted. Clinical trial information: NCT01993498 .