Quality of life of adolescent and adult survivors of childhood cancer in Europe—A systematic review

Advances in diagnostics and treatment of childhood cancer during the past few decades have substantially increased survival, resulting in a growing population of survivors of childhood cancer. Somatic and mental late effects of the cancer and the treatment may impact the quality of life (QoL). Previous reviews of QoL in survivors of childhood cancer have shown contradictory findings across studies and the majority of studies included have been based on data from North America and may not be directly comparable to a European setting. The aim of our study was to critically evaluate and summarise the latest evidence on the QoL of childhood cancer survivors in Europe and to identify survivors at particular risk. The eligible studies were published between 2008 and 2022, conducted in Europe and included participants who had survived at least 5 years after diagnosis of a childhood cancer. The main outcome of interest was QoL of survivors which was measured with validated qualitative and quantitative QoL questionnaires. A systematic literature search conducted in PubMed, EMBASE, PsycINFO and CINALH resulted in inclusion of 36 articles with a total of 14 342 survivors of childhood cancer. The majority of included studies found that childhood cancer survivors reported poorer QoL than comparisons. Female gender, treatment with haematopoietic stem cell transplantation and a brain tumour diagnosis were associated with lower QoL. With a growing population of childhood cancer survivors with many years ahead of them, targeted interventions and optimal follow‐up care are important to improve the QoL of survivors.

differences were small to moderate. Survivors of brain tumours, female survivors and survivors treated with haematopoietic stem cell transplantation were at particular risk. Targeted interventions and optimal follow-up care are important to improve the quality of life of childhood cancer survivors.

| INTRODUCTION
In Europe, approximately 15 000 children aged 0 to 14 years and 20 000 adolescents and young adults aged 15 to 24 years are diagnosed with cancer each year. 1 As a result of more effective treatment, the 5-year survival after childhood cancer has increased markedly over the past few decades and exceeds 80% in most European countries. 2,3 This substantial improvement in survival has resulted in a steadily growing population of childhood cancer survivors, who are, however, at increased risk of a wide range of somatic and psychiatric disorders, adverse mental late effects such as fatigue, neurocognitive sequelae (eg, impairments in memory), anxiety, depression, sleep disturbances and long-term socioeconomic consequences, such as lower educational attainment and unemployment. [4][5][6][7][8][9][10][11][12] These late effects may have a negative impact on the quality of life (QoL) of survivors. 13 For the sake of simplicity, the population of survivors who were diagnosed with cancer before the age of 21 will henceforth be referred to as childhood cancer survivors.
Previous reviews of studies of QoL in survivors of childhood cancer have shown that overall survivors are similar to comparisons with regard to physical, psychological and social domains of QoL but that subgroups of survivors have impaired QoL. Older age at diagnosis, longer time since diagnosis, female gender, being a survivor of a central nervous system (CNS) tumour or bone cancer, treatment with radiation therapy and certain socioeconomic factors such as lower educational attainment have been found to reduce QoL. [14][15][16][17] Lack of comparability among studies in these reviews obviates firm conclusions as populations were highly diverse. The study populations consisted, for example, of both children and adults, survivors and patients with current cancer and used both proxy and self-reported outcomes. Further, most of the studies in the previous reviews were of North American samples, and the results may not be generalizable to a European setting because of cultural differences and variations in healthcare systems. Therefore, we have focused on increasing comparability among studies by harmonising the population, which therefore only includes adolescent and adult survivors of childhood cancer in Europe.
The aim of this review is to critically appraise and summarise evidence from studies on the QoL of survivors of childhood cancer in Europe published since the latest relevant review. 17 A further aim is to identify factors related to poor QoL and survivors who are at particular risk.

| MATERIALS AND METHODS
The PRISMA guidelines were used for this systematic review. 18 The research strategy and eligibility criteria were determined, and then  Table S1. To find further studies, the reference lists of included articles were examined, and the latest citations were checked in the Scopus database.

| Study selection and eligibility criteria
As relevant studies published up to February 2008 were included in a previous systematic review, 17  This criterion was changed from 20 years during study selection to ensure inclusion of additional relevant articles. Studies of QoL in survivors under 15 years of age have often had proxy respondents rather than self-reports of QoL by survivors, which may emphasise physical rather than psychological problems. Therefore, we set the minimum age for evaluation of QoL at 15 years. As the QoL of newly diagnosed individuals still undergoing treatment may differ from that of individuals who have been cured, we included only studies of children who had survived for at least 5 years after the initial diagnosis to exclude acute effects of diagnosis and treatment. Intervention studies, reviews and studies conducted to validate or describe QoL questionnaires were excluded.
The screening and selection process was performed using the software Covidence. Two independent researchers (PAL, CP) assessed each article from its title and abstract and then extracted and screened full-text articles according to the eligibility criteria. Discrepancies between the two researchers were discussed and resolved by agreement. The data extracted (PAL, CP) from the articles included the first author's name, year of publication, country, study design, cancer type, sample size, age at diagnosis and evaluation, follow-up since diagnosis, period of diagnosis, comparison group(s), questionnaires used, factors related to QoL and main QoL outcomes. All data extracted from articles is presented in Table 1 questionnaires and different types of comparison groups, it was not feasible to conduct a meta-analysis on all of the included papers.

| Quality assessment
As all the studies included were cross-sectional in design, the Appraisal Tool for Cross-sectional Studies (AXIS) was used to assess their quality. 55 Table 1, and the QoL instruments used in the studies are presented in Table 2. The assessment of study quality is shown in Table S2.
Three studies found no significant differences in perceived QoL 30,36,50 and five studies found that survivors rated QoL or subscales higher than a comparison group. 19 54 The five studies that found a higher QoL in survivors were based mainly on survivors of haematological cancers. 19,40,42,43,45 One of the three studies that found no difference in QoL was based on brain tumour survivors, 50 while the other two were based on survivors of various cancers. 30,36 Of the 23 studies that found a lower QoL in survivors, F I G U R E 1 Flow chart of study selection process. regular daily activities due to physical health). 25,31,33,34,38,39,44,46,53 Fewer studies found that the subdomain BP was coherent with worse QoL. 25,32,34,35,39,46 The size of the effect varied by subscale The main mental subdomains found to be coherent with worse QoL were SF 22,25,32,34,35,[37][38][39]46,48,51,52 and MH, 20,25,32,34,35,44,46,49,51 but several studies also found that VT 20  T A B L E 2 Instruments used to assess the quality of life in studies of childhood cancer survivors.

Instrument (ref.)
No. of studies in which used Domains covered SF-36 56 18 Physical functioning, role limitations due to physical problems, social functioning, bodily pain, general mental health, role limitations due to emotional problems, vitality and general health perceptions RAND-36 57 5 Physical functioning, role limitations due to physical problems, social functioning, bodily pain, general mental health, role limitations due to emotional problems, energy/fatigue and general health perceptions SF-12 58 4 Physical component score and mental component score

| Quality of life-related characteristics
The characteristics correlated with QoL were examined in 32 studies.
We categorised the findings as cancer and treatment, physical and mental and sociodemographic characteristics. Studies in which characteristics were significant can be found in Table 3.

| Cancer and treatment characteristics
Survivors' age at diagnosis was not correlated with QoL in most studies. 19,23,24,30,31,35,36,41,42,46,49,53 In four studies, however, older age at diagnosis was associated with lower QoL. 22 which could be the mechanism by which survivors experience worse QoL. This is supported by several studies included in this review, which showed that QoL was correlated with the physical late effects of the cancer and its treatment and also with mental health problems. 26,29,31,32,34,42,46 The findings with regard to the impact of certain cancers on QoL were ambiguous, but, as previously observed, 14 survivors of brain tumours and solid tumours tended to report poorer QoL than survivors of haematological cancers. 29,31,32 One study found that the mediating factors between survival from a brain tumour and worse QoL were psychosocial, cognitive and neurological late effects; physical late effects and passive coping were other mediating factors. 29 Female childhood cancer survivors more often experience reduced QoL. 19,28,30,42 The reasons may include the fact that females tend to discuss problems more openly than men, 76 reflect the findings for the general population 77 or reflect differences in tolerance of therapy. 16 Although most of the studies in this review found a lower QoL in survivors of childhood cancer than in population norms or comparisons, four studies found a positive effect. 19,42,43,45 Possible explanations include posttraumatic growth after defeating a lifethreatening disease, a shift in values and perspectives and acquisition of an increased ability to cope with problems. 76,78 Other proposed explanations are a desire to appear normal and selfdeception response bias (an unconscious tendency to deny problems and therefore underreport symptoms and exaggerate positive aspects). 79 The findings of this review imply that treatment regimens should be continually reassessed, that multidisciplinary care should be part of the follow-up and that further research should be conducted on specific cancers and specific treatments to attenuate potential adverse effects and improve QoL. Future research should also evaluate targeted interventions to prevent and mitigate both health and psychological problems related to the cancer and its treatment, which could positively affect long-term QoL. Actions, when warranted, include less invasive treatments, risk-targeted follow-up to prevent, detect and treat health problems and targeted training programmes to improve daily functioning.

| Strengths and methodological challenges
A major strength of this comprehensive systematic review is the methodological approach. Four relevant databases were searched for articles, and the reference lists of the studies that were included were reviewed to ensure inclusion of all studies. Another strength is that only studies with self-reported measures of QoL were included. In previous studies, it was found that proxies cite fewer psychological problems and more physical problems than self-reports by survivors. 80 A third strength is that the participants evaluated had survived for at least 5 years, so that the findings are likely to reflect actual long-term complications of the cancer and its treatment and not immediate con-

| CONCLUSION
The findings of this review indicate that survivors of childhood cancer report worse QoL than comparison groups and normative data, although the differences in scores were generally small to moderate.
Certain cancers such as brain tumours, treatment with HSCT and late effects appeared to be risk factors for poorer QoL. In addition, female survivors tended to report lower QoL. This systematic review of QoL in European survivors of childhood cancer provides clinicians with important knowledge about possible complications that require attention. Research on the mechanisms underlying poor QoL could help to prevent or alleviate problems, as procedures could be adjusted or changed according to the findings. As more children survive a cancer diagnosis and more challenges are identified, it will be crucial to determine the appropriate care for each child and to target those in greatest need.

FUNDING INFORMATION
Our study was supported by the Danish Childhood Cancer Foundation.