Predictors of health-related quality of life and adjustment to prostate cancer during active surveillance

Abstract

Background: Active surveillance (AS) is emerging as an alternative approach to limit the risk of overtreatment and impairment of quality of life (QoL) in patients with low-risk localised prostate cancer. Although most patients report high levels of QoL, some men may be distressed by the idea of living with untreated cancer.

Objective: To identify factors associated with poor QoL during AS.

Design, setting, and participants: Between September 2007 and March 2012, 103 patients participated in the Prostate Cancer Research International Active Surveillance (PRIAS) QoL study. Mental health (Symptom Checklist-90), demographic, clinical, and decisional data were assessed at entrance in AS. Health-related QoL (HRQoL) Functional Assessment of Cancer Therapy-Prostate version and Mini-Mental Adjustment to Cancer outcomes were assessed after 10 mo of AS.

Outcome measurements and statistical analysis: Multivariate logistic regression models were used to identify predictors of low (<25th percentile) HRQoL, adjustment to cancer, and a global QoL index at 10 mo after enrollment.

Results and limitations: The mean age of the study patients was 67 yr (standard deviation: ±7 yr). Lack of partner (odds ratio [OR]: 0.08; p=0.009) and impaired mental health (OR: 1.2, p=0.1) were associated with low HRQoL (p=0.006; area under the curve [AUC]: 0.72). The maladaptive adjustment to cancer (p=0.047; AUC: 0.60) could be predicted by recent diagnosis (OR: 3.3; p=0.072). Poor global QoL (overall p=0.02; AUC: 0.85) was predicted by impaired mental health (OR: 1.16; p=0.070) and time from diagnosis to enrollment in AS <5 mo (OR: 5.52; p=0.009). Influence of different physicians on the choice of AS (OR: 0.17; p=0.044), presence of a partner (OR: 0.22; p=0.065), and diagnostic biopsy with >18 core specimens (OR: 0.89; p=0.029) were predictors of better QoL. Limitations of this study were the small sample size and the lack of a control group.

Conclusions: Factors predicting poor QoL were lack of a partner, impaired mental health, recent diagnosis, influence of clinicians and lower number of core samples taken at diagnostic biopsy. Educational support from physicians and emotional/social support should be promoted in some cases to prevent poor QoL.