Health-related quality of life of child and adolescent retinoblastoma survivors in the Netherlands

Retinoblastoma (RB) is a malignant tumor affecting the retina and is the most common intraocular malignancy in children. In the Netherlands, the incidence is 1:17,000 newborns (approximately 10–15 new patients every year)[1]. The survival of children with RB has significantly improved to a 5-year disease-free rate of more than 90% in the western world [2]. RB can affect one or two eyes, with subsequent blindness or severe visual impairment. The hereditary form of RB (40% of cases) is mostly bilateral and those patients are treated with enucleation and/or external beam radiotherapy (EBRT), and/or thermo-chemotherapy. Patients with unilateral RB are mostly treated with enucleation; after enucleation a prosthesis is inserted.

Considering the high survival rate of RB patients and the severe impact of the late effects of RB and its treatment, it is important to evaluate health-related quality of life (HRQoL) of RB survivors. So far, little attention has been paid to this subgroup of pediatric cancer survivors. Many studies have reported on QoL of pediatric cancer survivors but, to our knowledge, only Sheppard et al. (2005) have assessed the general QoL of young RB survivors [3]. Their study showed that mothers experience compromised QoL in their children (8–16 years), particularly with regard to their physical and psychosocial functioning, as compared to population norms. However, no study has yet made use of child self-reports, nor were any health-related questionnaires applied. Children are able to report reliably on their own well-being and functioning if the questionnaire is appropriate to the child's age and cognitive level[4]. Several factors that may jeopardize the HRQoL of RB survivors include, cosmetic deformities due to enucleation or EBRT[5, 6], visual impairment[7, 8], enhanced risk of second primary tumors [9, 10] and, in heritable RB patients, the 50% potential for offspring with RB.

The aim of the present study is to investigate HRQoL of young Dutch RB survivors using a self-report and a proxy measure, and to evaluate potential effects of illness-related and demographic variables on HRQoL. Further, to compare HRQoL of the survivors with that of healthy matched individuals, controlling for age and gender. Finally, to compare the children's own evaluations of their HRQoL with that of their parents.

The present study assessed the HRQoL of a unique Dutch population of young RB survivors using the KIDSCREEN questionnaire. To our knowledge, this is the first study to examine HRQoL in RB survivors with both a child and adolescent self-report, and a parent proxy-report questionnaire.

Our results suggest that the perceived HRQoL of children and adolescents who survived RB is not substantially different from the HRQoL of the normal population. It is noteworthy, however, that RB survivors report better "moods and emotions" and that they consider themselves to be more autonomous than healthy children. Parents of RB survivors report their child's HRQoL to be comparable with that of healthy children. This latter result is only partly in line with our earlier study, in which adult RB survivors were shown to experience a relatively good overall QoL, but a slightly worse mental health compared with a population-based reference group [15].

The discrepancy between adults and younger survivors might be caused by the fact that child and adolescent survivors are probably less well informed and/or are less aware of the effects in later life of the hereditary form of RB, such as risk for offspring, and enhanced risk for second primary tumors. During the home visits, some parents of hereditary RB survivors reported difficulties in informing their child about these effects, and were uncertain about the timing and the way the message should be conveyed. Shankar et al. (2005)also suggest that young (aged 8–12 years) childhood cancer survivors are too young to have encountered some of the negative psychosocial impacts of cancer and its treatment [16].

Studies assessing HRQoL of survivors of childhood cancer have, in general, reported contradictory results [17]. Our findings are in line with studies of Shankar et al. (2005), Pemberger et al. (2005), Apajasalo et al. (1996), Gray et al. (1992), Langeveld et al. (2004) and Maunsell et al. (2006), which report that long-term childhood cancer survivors show comparable or even higher-than-average positive subjective rating of the various areas of HRQoL [16–22]. According to these general childhood cancer studies, the observed excellent HRQoL could be explained in two ways. First, following the theory of response shift it may be that (as a result of the experience of cancer) the internal standards, values or conceptualizations of HRQoL change [23]. The experience of surviving cancer might lead to a better appreciation of being alive and to considering that possible impairments are of less importance. Barakat et al. (2006) found that a majority of adolescent childhood cancer survivors and their parents reported post-traumatic growth (PTG) [24]. Greater perceived treatment severity and life threat was associated with PTG. Diagnosis of cancer after age 5 years resulted in more perceived benefit and greater posttraumatic stress symptoms for adolescent survivors. Second, coping with the stress of the long-term effects of childhood cancer may have enhanced certain qualities of the survivors making them better able to cope with adversity in their lives.

It is conceivable that some of above-mentioned explanations may also apply to child and adolescent RB survivors. However, because this latter group often had treatment at a very young age, the time since treatment is long and the period of treatment in most cases was relatively short compared with most other forms of childhood cancer. This implies that, for this group, there may be less opportunity for the development of compensating abilities. In contrast to our results, other studies have reported a lower HRQoL for childhood cancer survivors in comparison with a healthy population [25–27]. Clearly, further research is necessary to assess the plausibility of the various explanations related to this group.

With respect to the survivor's own perceptions of their HRQoL, the following findings are of particular interest: especially adolescent RB survivors feel themselves to be more autonomous than healthy adolescent children, whereas younger RB survivors do not yet share this feeling. Younger RB survivors perceive themselves to be happier and as having closer relationships with their parents than children who did not experience RB. On closer inspection, increased ratings of HRQoL in RB survivors are mainly traceable to perceptions of adolescent girls who report themselves to be in a better mood, more autonomous, better at school and financially more independent than girls in the Dutch reference group. Thus the good HRQoL reported by RB survivors during childhood and early adolescence remains true, despite the fact that adolescents are generally more able to reflect on their own functioning and appearance. Indeed, in our group we found an overall significant negative effect of age regarding psycho-social well-being, possibly even concealing the remarkable increase of HRQoL in girls.

Our RB survivors with a normal vision reported better physical well-being than visually impaired RB survivors. This result is in line with a recent methodological study by Birch et al. (2007) which also confirmed the common expectation that children with more severe visual impairment experience poorer competence than those with only "unilateral" impairment [28]. In our study, parents of visually impaired RB survivors reported that their child had a higher level of autonomy than did parents of children with normal vision. Relatively good psychological adjustment in the more affected children may be caused by increased parental attention, as was observed in a sibling study [29].

With respect to parents' perceptions of their child's HRQoL, it is noteworthy that these are very different from the child's own perceptions. In general parents do not experience any difference in the HRQoL of their own child as compared to matched groups. Parents of young boys observed their son to be even less autonomous than their male peers. Within the RB group, parents of visually impaired RB survivors rated their child to be more autonomous than parents of RB survivors with normal vision. This finding may touch on a compensatory mechanism; however, we cannot exclude that visually impaired children undergo a different learning process whereby they learn to function better autonomously than children with normal vision. These findings are partly in concordance with Sheppard et al. (2005) who found that mothers reported lower levels of HRQoL for their children and adolescents treated for RB compared with population norms [5].

Some limitations of the present study should be addressed. First, the small number of survivors limits the statistical power of the study. Second, it is conceivable that some of the non-participating RB survivors experienced a different HRQoL. It is not clear, however, whether this would result in better or worse results. For example, patients who feel good might disregard the importance of the study and, on the other hand, parents of non-participating survivors may have refused participation to avoid their child being confronted with their disease again, which might suggest serious concern of the parents, or worse coping strategies and a poorer QoL. Third, the HRQoL of the survivors was only partly explained by the factors that were investigated in the present study. Other explanatory factors, such as coping and family functioning, may also be of importance; future studies should explore these factors. Indeed one would expect coping skills and HRQoL to interact, but unfortunately it was beyond the scope of this study to include such measures. Fourth, the results might be influenced by the choice of a general HRQoL questionnaire. In some specific cases our clinical impression seemed to be in contrast with the positive results of our study. An example we sometimes encountered in clinical practice is that RB survivors often have a different facial appearance due to the treatment; this may often lead to bullying and staring. In this study we found it important to compare our data with a healthy Dutch reference group, but a general HRQoL instrument (such as KIDSCREEN) only measures broad areas of HRQoL and may not identify such issues specifically associated with RB. In future studies it is advisable to use an RB-specific instrument or a vision-related HRQoL instrument besides a general HRQoL instrument. Unfortunately there is a shortage of such instruments, which in itself presents a challenge for future research.

In conclusion, child and adolescent RB survivors report a good QoL compared with the Dutch reference group. It is noteworthy that perceptions of HRQoL as reported by parents and by children are very different. Parents judge the HRQoL of their child to be relatively poorer. Although the overall results are reassuring, other aspects of HRQoL that may have more specific relevance, such as psychological factors or coping skills, should also be explored in the future.