Impact of childhood leukemia and its treatment on Nordic children and families : the consent process, toxicity, and quality of life

Abstract

Background: The survival rate in acute lymphoblastic leukemia (ALL) has improved remarkably over the last half-century and is now reaching above 90%. The treatment is lengthy, however, and with a high risk of short- and long-term side effects. The advances in treatment outcome have underscored the importance of addressing not only cure, but also the potential side effects of the treatment, as well as quality of life. The Nordic Society of Paediatric Haematology and Oncology (NOPHO) ALL2008 protocol was used in the Nordic and Baltic countries between 2008-2019. Treatment duration was 2.5 years, and there were 3 randomized clinical trials (RCTs) within the protocol. RCTs are essential in improving treatment outcomes, but the process of being informed and asked to decide on RCT participation may put further stress on the families.

The aim of this thesis was to investigate the impact of the treatment period on the children and their families by assessing parents’ experiences of the informed consent process for the RCTs, by evaluating health-related quality of life (HRQOL) in the children and their parents, and by assessing how parents perceived the different treatmentrelated side effects observed in their children.

Methods: Families of patients aged 1-17.99 years at diagnosis in Sweden, Finland, and Denmark who had been treated according to the NOPHO ALL2008 protocol and were alive and in first remission ≥6 months after end of therapy, without any secondary malignancy, were eligible. Patients with Down syndrome, mixed phenotype leukemia, or stem cell transplant in first remission were excluded for substudies I-III, but their parents were eligible for substudy IV. Three different questionnaires were sent: First, a studyspecific questionnaire that was developed by the authors in collaboration with 7 families of children previously treated for ALL in Stockholm through a face validation process. It contained questions on socioeconomic factors before and after treatment, sibling support, toxicity and impact of treatment, and how parents perceived the process regarding being asked to consent to RCT participation. The second questionnaire was the HRQOL-instrument PedsQL™ 4.0 Generic Core Scales (self and proxy-report) for the child’s HRQOL, and the third was the SF-36 for parents’ HRQOL. Eligibility was crosschecked with local nurses at each center before sending the questionnaires to patients and their families. Data were collected between 2013-2019.

Results: Overall, families of 320 children participated in at least 1 of the substudies. In substudy I (n=483 parents of 279 children), on the informed consent process for the RCTs, we found that parents in general reported the consent process as satisfactory, with sufficient time and information to make a well-informed decision and that they did not feel pressured to participate. Those who declined study participation reported a more negative experience overall, with significantly less confidence in study design and less satisfying process that those who consented. More than a third of all parents, and over half of parents who had refused participation, felt that it was burdensome to have to make the decision. Two-thirds of parents in general, and one-third of those with children 8 years or older, reported that their child had not been involved in the decisionmaking process. In substudy II (n=299 children; 276 self-reports and 270 proxy-reports), children over 8 years reported HRQOL similar to that of a Finnish reference population, except lower scores in the school functioning domain for high-risk patients. Children 5-7 years old had notably lower scores in all domains, which may in part be due to the different construct of the PedsQL™ questionnaire for this age group specifically (with 3, instead of 5, levels on the response scale for their self-report). Parent-proxy reports were also lower than reference data. Parent-reported toxicity in the study-specific questionnaire was associated with lower total and physical HRQOL scores in adjusted models for all ages in both self- and proxy reports, and for parent-proxy reports also lower psychosocial HRQOL. Substudy III investigated how parents perceived treatment-related side effects during ALL treatment, with focus on specific side effects from vincristine, corticosteroids, pegasparaginase, and maintenance therapy. Parents of 307 children (296 mothers, 244 fathers) responded. For vincristine and corticosteroids, more than a third rated the overall negative impact as high. Walking difficulties, muscular weakness, and pain, and appetite and mood swings, were the most common and severe side effects for vincristine and corticosteroids, respectively. For peg-asparaginase and maintenance therapy, symptoms were less frequent and the overall impact reported by parents was less than for vincristine and corticosteroids. For toxicities also reported in the NOPHO ALL2008 database, e.g., thrombosis and pancreatitis, parent reports were similar with database reports. Although parents in general reported that their child had been negatively affected to a high extent overall during treatment, they reported that their child was affected to a lower, or similar, extent when asked to compare with other children with ALL. In substudy IV, parents’ HRQOL were assessed using the SF-36 questionnaire. Data from 526 parents of 310 children showed that the participating parents had significantly lower scores than the reference population on both physical and mental component summary measures that surpassed minimal clinically important difference for the mental component summary. Mothers had lower mental HRQOL than fathers as well as higher absence from work. Higher mental HRQOL scores were also associated with living in Denmark or Finland (compared to Sweden). When comparing HRQOL within couples living together, correlations were weak to moderate.

Conclusion: ALL treatment impacted the children and their families to a large extent. Parents reported high levels and severe impact of treatment-related side effects, and children with higher parent-reported toxicity compared to other ALL patients had lower HRQOL. This determinant was more important for child’s HRQOL than i.e., risk group or socioeconomic factors, and was not seen for toxicity in the NOPHO ALL2008 database. Children 8 years and older otherwise reported their HRQOL at similar levels as the reference population, while parent-proxy reports, and younger children (5-7 years), reported lower HRQOL. The parents' own HRQOL was negatively affected, especially the mental domains and especially in mothers, which may indicate a need for extra support. For the informed consent process, parents were generally satisfied with the process overall, although a high number of parents thought it was burdensome to make the decision, especially among those who refused participation. Fewer than expected of the school-aged children were involved in the decision-making process, and efforts to improve children’s involvement in this process should be considered. The work done within this thesis highlights the importance of including patient/parentreported outcomes, not least in toxicity reporting, and of the task to keep striving to minimize complications from treatment in future ALL protocols. It also shows some of the burden that parents experience, and calls for awareness that RCT decisions put a stress on families, even when declining participation, and of parents' high levels of altruism, confidence, and trust.