Health-related quality of life and its determinants during and after treatment for paediatric acute lymphoblastic leukaemia: a national, prospective, longitudinal study in the Netherlands

Abstract

Objectives: Health-related quality of life (HRQoL) is impaired in paediatric patients with acute lymphoblastic leukaemia (ALL). Over the past decades, ALL treatment has successfully been adjusted to the risk of relapse, which is now reflected by the stratification of patients into three risk groups who receive treatment of differing intensities. This study is the first to evaluate the longitudinal course of HRQoL in light of these adjustments and identify determinants of HRQoL.

Design: Two prospective, national cohort studies (add-on studies within the two most recent treatment protocols for children with ALL (ALL-10 and ALL-11)).

Setting: Dutch paediatric oncology hospitals between October 2006 and October 2009 (ALL-10) and between August 2013 and July 2017 (ALL-11).

Participants: Patients with ALL (2-18 years) are treated according to the ALL-10 or ALL-11 treatment protocol. Patients treated according to the ALL-10 protocol only completed a cancer-specific QoL measure and patients treated according to the ALL-11 protocol completed both a cancer-specific and generic QoL measure (see below).

Outcome measures: HRQoL, assessed with parent-proxy questionnaires (PedsQL Generic and Cancer module) within the first 5 months (T0), at 1 year (T1), 2 years (T2) and 3 years (T3) after diagnosis. The proportion of patients with clinically relevant generic HRQoL impairment was compared with healthy norm values. Multivariable mixed model analyses were used to evaluate the development of HRQoL over time and its medical and sociodemographic determinants (collected on enrolment).

Results: Of the ALL-10 cohort, 132 families participated and of the ALL-11 cohort, 136 families participated (268 total). Thus, cancer-specific HRQoL assessments were available for 268 patients (median age 5.3 years (IQR 6.15), 56.0% boys, 69.0% medium-risk ALL), and generic HRQoL assessments for 136 patients (median age 4.8 years (IQR 6.13), 60.3% boys, 75.0% medium-risk ALL). Generic HRQoL improved between timepoints T0 and T3 (total score B 16.1, 95% CI 12.2 to 20.1, p<0.001), but did not restore to normal 1 year after the end of treatment: 28.0% of children remained impaired compared with 16% in the general population (p=0.003). Cancer-specific HRQoL generally improved from T0 to T2 (Pain B 11.3, 95% CI 7.1 to 15.5; Nausea B 11.7, 8.4 to 15.1; Procedural Anxiety B 19.1, 14.8 to 23.4; Treatment Anxiety B 12.8, 9.5 to 16.0; Worry B 3.5, 0.6 to 6.3; Communication B 8.5, 5.0 to 11.9; all p<0.001 except for Worry (p=0.02)), while Physical Appearance and Cognitive Functioning remained stable. Higher treatment intensity and experiencing pain or simultaneous chronic illness were associated with lower HRQoL over time for multiple subscales.

Conclusions: HRQoL impairment is prevalent during and after ALL treatment. Patients with standard-risk ALL and reduced treatment intensity have better HRQoL than patients in higher risk groups. Systematic monitoring of HRQoL is of utmost importance in order to provide timely psychosocial interventions and supportive care.

Keywords: Leukaemia; ONCOLOGY; Paediatric oncology.

Figure 1

Schematic overview of ALL treatment and study assessments. *End of treatment for patients with IKZF-1 deletion. **Additional assessment with dexamethasone for medium risk patients. ***Includes stem cell transplantation for the majority of patients. ****Assessment not conducted for high risk patients. ALL, acute lymphoblastic leukaemia.

Figure 2

Longitudinal development of generic HRQoL scores. *p<0.05. HRQoL, health-related quality of life; PedsQL, Pediatric Quality of Life Inventory.

Figure 3

Longitudinal development of cancer-specific HRQoL scores. *p<0.005. HRQoL, health-related quality of life.

Figure 4

Proportion of patients reporting problems (‘never’ to ‘always’) on the PedsQL Cancer scales. PedsQL, Pediatric Quality of Life Inventory.