Quality of life among people with eye cancer: a systematic review from 2012 to 2022

Abstract

Background

Eye cancer is a serious eye disease that threatens patients’ lives. In the past decade, there have been more and more studies on eye cancer. From the recently published eye cancer literature review, it can be seen that the two most popular research hotspots are retinoblastoma (RB) and uveal melanoma (UM) [1, 2]. Although several studies have assessed QOL in different types of eye cancer patients, a study that synthesizes the factors influencing QOL in eye cancer patients is yet to be undertaken. This study aimed to review and evaluate the literature related to the QOL of RB and UM survivors, and provide a synthesis of the current evidence on the impact of the two types of eye cancer on the overall QOL of patients.

Methods

Eight databases (APA Psych Articles, CINAHL Complete, Health Source: Nursing/Academic Edition, MEDLINE Complete, Scopus, Cochrane Library, PubMed, and Registers (Clinicaltrials.gov.)) were searched between January 2012 and December 2022 for English, peer-reviewed quantitative original studies within this review. All publications were screened using the Preferred Reporting Items for Systematic Review and Meta-Analyses reporting guidelines. The methodological quality of the reviews was assessed using the Joanna Briggs Institute Critical Appraisal Checklists. The findings were summarised and tabulated accordingly.

Results

Seventeen articles were analysed. Among them, 14 articles on patients with UM, and three articles on patients with RB using 18 different types of measurement tools were included. Eight researchers claimed that the overall QOL of patients with eye cancer was better than or similar to that of the general healthy population. However, nine studies indicated that these patients had poorer QOL than others. Many factors affect QOL, including treatment, sex, and age.

Conclusion

This systematic review identified the QOL levels and several factors that influence the QOL of ocular cancer patients worldwide, due to the variability in quality of the studies, it also showed the need for further research to assess factors affecting long-term QOL outcomes in RB and UM survivors. Simultaneously, it clarified the necessity and importance of developing standardized and complete assessment tools to compare QOL in different countries. Early interventions can be developed to improve the survivors’ QOL by identifying potential deficits in specific areas.

Background

Quality of life (QOL) is a multidimensional construct that includes physical, emotional, functional, social, and family well-being [3]. It is considered as an important patient-reported outcome indicator that helps promote patient-doctor communication, detect symptoms, and influence medical decision making [4]. Furthermore, QOL is a highly subjective, dynamic process affected by changes in lifestyle and events [5]. Ocular tumours are serious ophthalmic diseases that threaten eyesight, QOL and morbidity, and mortality [6]. The two most prevalent eye cancers are retinoblastoma (RB) and uveal melanoma (UM) [7]. RB is a malignant eye cancer that occurs during childhood, accounting for 2–4% of childhood malignancies [1], with an incidence rate of 1/15,000–20,000 [8], and nearly 70% of cases are diagnosed within 2 years of age [9]. Globally, there are 8,000–9,000 new cases of RB being diagnosed each year [10]. Over the past 40 years, the incidence of RB in the United States (US) has been stable [11]. There is no difference in the incidence of RB according to race, sex, and region, but the survival rates of patients with RB in different countries and regions differ significantly [12]. In the US, the survival rate of RB patients reaches 95%, but in some underdeveloped countries, especially some countries in East Africa, the survival rate of patients with RB is as low as 30% [13,14,15,16]. UM is the most common intraocular malignancy in adults, affecting 2–8 cases per million people per year in Caucasians in Europe [7]. Furthermore, regardless of treatment, the 10-year metastasis-free and overall survival rates for UM-medium-sized tumours are 87% and 65%, respectively [17]. Once metastasis occurs, the average survival time of patients is 3–4 months, and the 1-year mortality rate is 80% [18]. Up to 50% of patients die from metastases within 10 years of UM diagnosis [19]. Furthermore, regardless of the treatment type, the survival rate of patients remains unchanged [20, 21]. The traditional treatments for RB and UM include enucleation surgery, chemotherapy, radiotherapy, local treatment, gene therapy and vitreous surgery [2]. Enucleation has been the standard treatment for eye cancer [22, 23], with 5-year survival rate post-surgery from 17 to 53% [24]. Although an increasing number of patients [23,24,25] undergo enucleation therapy, their prognosis remains poor. Approximately 35% of patients undergo enucleation [26]. Patients who are older, have more advanced tumours, and have a higher risk of metastasis often choose enucleation, which impacts health conditions [26], especially psychological trauma [12]. Moreover, enucleation is a destructive and disfiguring surgery that seriously impacts a patient’s physical and mental health [27]. In addition, postoperative vision loss can cause problems such as distance perception, and prosthetic eyes may result in irritation, discomfort, pain, and unsatisfactory appearance [28]. With the advancement of medical treatments, patients can choose between brachytherapy and proton beam radiotherapy [29, 30]. Brachytherapy is a type of radiotherapy wherein the treatment region is surrounded by a sealed radiation source. UM requires the placement of a plaque over the malignancy inside the eye using sutures from outside the eye. Although proton beam radiotherapy can preserve the eyeball with useful vision remained [21], it also increases the risk of neovascular glaucoma (44.8%), radiation retinopathy (25.4%), retinal detachment (16.8%) and local tumor recurrence (12.5%) [31].

Nearly all survivors are at an increased risk of secondary malignancies, yet concerns about the visual and genetic components of their disease remain unabated [32]. Irrespective of enucleation therapy, they are often left with monocular vision, which negatively affects motor processing, judging distance, depth perception, and increased the risk of visual dysfunction, cataracts, and severe hearing loss [33]. However, little is known about the potential impact of these long-term sequelae on the QOL of survivors. Based on these insights, QOL is a key factor in selecting an appropriate management plan for these patients [34]. Life-threatening cancer experiences and adverse effects of chemotherapy, radiotherapy, surgery, and enucleation may potentially influence the QOL of survivors [35]. Measuring QOL is critical for determining the extent to which eye cancer and its treatment impact daily life [36, 37]. Several studies have explored the QOL of patients after treatment. However, researchers hold different views on this topic. Therefore, this systematic review aimed to evaluate the literature related to the QOL of adult survivors of RB and UM, by providing a synthesis of the current evidence investigating the impact of these two types of eye cancer on the overall QOL of patients. Furthermore, this review can provide advice on further studies on these patients and, measures to improve their QOL.

Methods

The Preferred Reporting Items for Systematic Review and Meta-Analyses (PRISMA) reporting guidelines were followed [38]. At the beginning of the study, a protocol was developed and registered with PROSPERO (CRD42022283279).

Search strategy

The search strategy was developed by the authors with the support of two librarians at Munster Technological University. The search strategy was tested using an iterative process before finalisation (Table 1). Eight databases were searched: APA Psych Articles, CINAHL Complete, Health Source: Nursing/Academic Edition, MEDLINE Complete, Scopus, Cochrane Library, PubMed, and Registers (ClinicalTrials.gov.). The reference lists of selected articles were checked for potentially relevant articles. The initial search was conducted on 19 May,2022, and updated on 31 December, 2022.

Table 1 Search strategy table

Eligibility criteria

Studies that investigated the QOL of patients with UM or RB only (> 18 years old), regardless of whether they were living at home or in any care setting were eligible for inclusion. Additional inclusion criteria were quantitative studies published in English between January 2012 and December 2022.

Study selection

All records were imported into Endnote X9 and duplicates were removed. Subsequently, the records were exported to Rayyan [39], where two reviewers (Y.H. and C.O.B.) independently completed the title and abstract screening based on the eligibility criteria outlined above. Following this step, the full-text articles were identified and read by the reviewers to determine their final eligibility. All eligible patients were retrieved and the full text was added to Rayyan [39]. Two reviewers (Y.H. and C.O.B.) read each article to determine eligibility for inclusion. Disagreement between the two reviewers was resolved by a third reviewer (D.L.).

Data extraction and risk of bias assessment

Two reviewers (Y.H. and Y.G.) independently extracted the data into an Excel file. The extracted data consisted of the study and participants’ characteristics (Table 2). The Joanna Briggs Institute critical appraisal checklists (for cross-sectional and cohort studies, as appropriate) were used to evaluate the risk of bias and quality of each article. This review was completed independently by two authors (Y.H. and C.O.B).

Table 2 Summary of the QOL outcomes by treatment modality

Data synthesis and analysis

A narrative synthesis of findings of all studies is presented below.

Results

The original search identified 11,857 articles. After removing duplicates, 6,282 titles and abstracts were screened for eligibility, with 37 remaining for a full-text review. Finally, 17 studies were included in the systematic review (Fig. 1).

Fig. 1

PRISMA flow diagram. Search 1* = APA Psych Articles, CINAHL Complete, Health Source: Nursing/Academic Edition, MEDLINE Complete. Total: 11,857 (19/05/2022)

Study characteristics

Most of the included studies (15), investigated UM [5, 17, 28, 30, 37, 40,41,42,43,44,45,46,47,48] with three studies on RB [16, 33, 49]. Among the included articles, one [49] study was carried out in a low- or middle-income country (China) and 16 studies were conducted in high-income countries: the US (9) [16, 17, 28, 30, 33, 40,41,42, 47], Germany (4) [37, 43, 46, 48], Ireland (1) [5], Israel (1) [44], and the Netherlands (1) [45]. In relation to study design, there are ten cohort studies [28, 30, 33, 37, 40, 41, 44,45,46, 48], five cross-sectional studies [5, 16, 17, 47, 49], and two longitudinal studies [42, 43]. The characteristics of the included studies are summarised in Table 2.

Patient reported outcomes of QOL - assessment tools

Eighteen different assessment tools were used to measure patients’ QOL across all included studies (Table 2). Eight articles used more than two different tools to measure outcomes [16, 30, 41, 42, 45,46,47, 49]. The most commonly used tools were Quality of Life Questionnaire Ophthalmic Module (QLQ-OPT30, measured the QOL for UM patients) [5, 17, 28, 30, 41, 42, 44, 46] and the Core Quality of Life questionnaire (QLQ-C30, measured the QOL for patients with all kinds of cancer) [5, 17, 44,45,46] authored by the European Organization for Research and Treatment of Cancer (EORTC); the Functional Assessment of Cancer Therapy-General (FACT-G, measured physical, social, emotional, and functional well-being domains of QOL in patients with cancer) [30, 40,41,42], National Eye Institute Visual Function Questionnaire (NEI-VFQ, measured the dimensions of self-reported vision-targeted health status that are most important for patients with chronic eye diseases) [33, 45, 47], Hospital Anxiety and Depression Scale (HADS, measure anxiety and depression in a general medical population) [30, 40,41,42, 49] and 12-Item Short Form Survey (SF-12, assessed the impact of health on an individual’s everyday life) [37, 43].

The overall QOL level

The findings reported in the included studies indicated variations in the QOL of eye cancer survivors across studies. Six studies showed that the overall QOL scores of patients with UM or RB were higher than or similar to those of the healthy general population [5, 16, 17, 40, 42, 49]. However, 11 studies indicated that these patients had a poorer QOL than the healthy general population [28, 30, 33, 37, 41, 43,44,45,46,47,48]. Three studies investigated QOL in patients with RB [16, 33, 49], two of which concluded that the overall QOL was better than that in non-cancer controls [16, 49], and one concluded that QOL was affected by vision-targeted HRQOL in some aspects; for example, subdomains of visual function such as peripheral vision, especially in bilateral survivors of disease (P < 0.001) or with enucleation (P = 0.002) [33]. There were 14 articles investigating the overall QOL of UM patients, 4 out of 14 reported that there is no difference between QOL of these patients and the general population [5, 17, 40, 42], 6 out of 14 reported that the QOL was affected and decreased [30, 37, 41, 43, 44, 47], and 4 out of 14 indicated that the QOL decreased first and then there was no difference with time [28, 45, 46, 48]. Six articles [28, 37, 40, 45, 46, 48] noted poorer QOL at the start of treatment but at the 3 [46], 6 [28, 40],12 [28, 40, 48], 24 [28, 40], 36 [37] and 48 months [45] follow-ups, there were no differences when compared to the general healthy population. In contrast, Frenkel et al. [44] suggested that treatment has a greater impact on the patient’s life during the first 2 years than later, however this was not statistically significant (P = 0.073); the other QOL domains were not related to the post-treatment duration. Only in the cohorts of van Beek et al. [45] and Gollrad et al. [46, 48], pretreatment status was assessed. In addition, van Beek et al. [45] consistently reported that anxiety problems improved in patients with UM within 2 years after primary treatment compared to pretreatment assessments [45]. They also found that from 2 months to 3 years after treatment, the role functioning score increased (P = 0.005) in patients who had undergone irradiation, and there was a decline in physical functioning (P = 0.035), insomnia (P < 0.001), and anxiety (P < 0.001), whereas increases in overall pain (P = 0.023) and emotional functioning (P < 0.001) were observed 1 year posttreatment [45]. Decrease in physical functioning (P = 0.007), role physical functioning (P < 0.0001), and improvement in mental health (P = 0.023) at 1- and 2-year follow-ups after stereotaxic CyberKnife radiosurgery were demonstrated by Klingenstein et al. [43]. In Gollrad et al. [46], subsequent emotional and social functioning improvement after treatment were similar to those of previous studies that improved overtime.

Overall, RB patients have better QOL scores than UM patients, with two out of three studies [15, 49] showing that RB patients had scores similar to general population, thereby reporting fewer QOL-related problems. However, the QOL in patients with UM is variable, with inconsistent results across studies. Only 6 of 14 studies [5, 17, 28, 37, 40, 42] indicated that the QOL was not different or better than that of the general population.

Age and sex differences in QOL

Ten studies reported the effects of age and sex on QOL [5, 17, 30, 33, 37, 40, 43, 46, 47, 49]. Younger female patients were more likely to have problems that affected their QOL or sub-items [5, 17, 37, 43, 46, 49], and higher anxiety [30, 40, 46] (P < 0.001). These differences can be classified into physical and mental outcomes. As for physical outcomes, Gollrad et al. [46] found that younger patients’ global health was worse (after the clip surgery procedure; P = 0.021), with more headaches (pretreatment and final assessment; P < 0.04, post-proton treatment (P = 0.006). Scannell et al. [5] also found that younger (12–54 years old) female patients were more likely to report headaches (P < 0.0005) and reading difficulties (P = 0.042). Similarly, Klingenstein et al. [37] found differences in physical functioning according to age and sex (all P < 0.05). Barker et al. [17] also found that age (P < 0.01) and sex (P = 0.03) influenced on physical functioning scores. Klingenstein et al. [43] reported that physical functioning decreased significantly in male survivors (P = 0.030), and role physical functioning decreased significantly in female survivors (P = 0.017). Damato et al. [30] also reported that women’s physical well-being improved in females. However, Wiley et al. [47] found that sex and age were not associated with role difficulties (P > 0.05). Friedman et al. [33] showed that age and sex did not affect NEI-VFQ-25 scores. Regarding mental outcomes, Gollrad et al. [46] found that only female patients had a significant decrease in QOL between pretreatment and post-surgical assessment (P < 0.001), with lower emotional functioning at all valuation points (pretreatment: P = 0.042, post-surgical assessment: P = 0.037,post-proton treatment: P = 0.014, final assessment: P = 0.004), and higher anxiety (P < 0.001). Barker et al. [17] also found that role functioning (P = 0.01) and emotional functioning (P < 0.01) were related to sex. Similarly, Feng et al. [49] found that women were more susceptible to the disease (P = 0.031) and more concerned about their appearance (P = 0.041). Further, Klingenstein et al. [43] reported that mental health in male survivors improved after 1 year (P = 0.042), but there was no difference at 2 years (P = 0.16). Hope-Stone et al. [40] assessed anxiety levels 2 years after treatment and found that younger patients were more anxious (P < 0.01). However, Wiley et al. [47] found that sex and age were not associated with mental health or fear of recurrence (P > 0.05). Similarly, Damato et al. [30] mentioned that women were more anxious but their emotional well-being was better. Klingenstein et al. [37] found that there were differences in role-emotional scores by age and sex (all P < 0.05), and in mental health scores by sex (P = 0.021), but they did not mention the specific differences.

Treatment differences in QOL

Twelve articles [5, 28, 30, 33, 37, 40,41,42, 44, 45, 47, 48] explored the relationship between the treatment received and subsequent QOL. Table 2 summarises the QOL outcomes reported by treatment in each included study. Among 12 comparative studies, three reported no overall difference in QOL between the different types of treatments [5, 41, 42, 47]. Three of 12 studies reported a significant difference in at least one QOL subdomain for one treatment modality compared to other modalities [28, 30, 33, 37]. Three of the 12 studies described no overall differences in long-term follow-up but noted some significant differences in QOL in specific functional domains (e.g. vision problems) [44, 45, 48], anxiety [40, 45], depression [40, 45], and short-term physical functioning [45].

Ten studies indicated that enucleation affected QOL [5, 28, 30, 37, 40,41,42, 44, 45, 47]. Klingenstein et al. [37] compared the QOL of patients with UM treated with CyberKnife or enucleation with that of the general population and found that radiation did not affect QOL, whereas enucleation decreased physical and emotional parameters, especially those associated with physical functioning (P = 0.0063), emotional role (P = 0.012), physical role (P = 0.043) and body pain (P = 0.037). Hope-Stone et al. [28] reported that patients with UM that had undergone enucleation faced more and worse functional problems at 6 months, which diminished at 12 and 24 months (P = 0.020); however, QOL did not differ in terms of driving difficulties (P = 0.694), eye irritation (P = 0.281), headaches (P = 0.640), appearance problems (P = 0.187) and fear of recurrence (P = 0.899). Damato et al. [30] reported that patients who underwent primary enucleation had worse QOL. However, as mentioned in their subsequent study [41], QOL among the different treatment modalities did not differ, and patients’ QOL recovered to a level similar to that of the general population within 6 months, which is in agreement with Hope-Stone et al. [40], who reported that there was no difference in any measure of QOL, regardless of enucleation. However, Scannell et al. [5] and van Beek et al. [45] reported no difference in QOL between patients who underwent enucleation and those who did not (89.3 vs. 89.2, respectively; 78.8 [fractionated stereotactic radiation therapy] vs. 78.3 [enucleation]). Frenkel et al. [44] also demonstrated that the general QOL did not differs at initial treatment, but those who underwent enucleation had a lower eye-related QOL than those who received brachytherapy (P = 0.019) which is in line with Wiley et al. [47], who claimed that compared with patients receiving brachytherapy, patients undergoing enucleation had more role difficulties. Brown et al. [42] found that, regardless of the treatment modality, emotional QOL increased in patients with UM at 6, 12, and 24 months after diagnosis.

Discussion

Despite the differences in QOL among patients with eye cancer, insufficient research has been conducted, and only 17 relevant studies have been published in past ten years. The included studies used 18 different kinds of questionnaires, and the proportion of UM-specific assessment tools (OPT-30) among patients reached 57.14% (8 of 14) [5, 16, 28, 30, 41, 42, 44, 46]. In addition to the core QOL questionnaire, 9 of 17 studies [16, 30, 40,41,42, 45,46,47, 49] addressed important aspects such as anxiety, depression, fears, concerns, and financial burden. However, with follow-up from baseline [45, 46, 48] to a maximum of 295 months [44], it was reported that treatment did not impact the survival rate of patients, but the QOL differed over time; therefore, a longitudinal assessment using the same measurement of QOL is essential [4, 50] and has become an important focus of clinical decision-making [5, 28, 37, 41,42,43,44,45,46]. Researchers have also noted that QOL is affected by various factors, such as age, sex, treatment, and general health.

Many factors could have caused bias among the 17 articles. As for the sampling distribution firstly, because of age, sex, social factors, tumour size, and location, it is immoral to assign patients to randomised controlled trials [5, 28, 30, 41, 45]. Secondly, as shown in Table 2, although the proportion of men and women included is roughly equal, the sample size varies greatly, ranging from 91 [43] to 1596 [30, 41] people. Some studies with small samples may lack the statistical significance of the data. However, owing to the low incidence and high mortality of this cancer, and the differences in follow-up time between different studies also pose challenges to the data collection of large-sample and long-term studies, therefore in the absence of sufficient time, researchers will obtain a small sample size [28, 44, 48]. Thirdly, only three articles had baseline data [45, 46, 48], which meant that the other studies were unable to have internal comparisons, because they did not know the original status of these patients. Fourthly, fewer enucleation patients were involved, which may be because patients received radiotherapy need more regular follow-up treatment, therefore easier for recruitment [30]. Similarly, Ford et al. [16] excluded disabled survivors, which is a selection bias. Simultaneously, Hope-stone et al.’s [28] cohort had a high loss of follow-up rate of up to 40.7%, and the other research conducted by them [40] also had a loss rate of 17%, which is similar to Feng. et al. [49], whose loss of follow-up was 30.2%, which means that some patients failed to complete the entire follow-up period and withdraw from the study, resulting in higher loss to follow-up rate, the result may not represent the real conditions of the survivors. Moreover, the differences in outcomes between countries may be related to different medical and social support systems, for example, the included studies come from 9 cities in 6 different countries, with different inclusion and exclusion criteria and regional differences, except for Klingenstein et al. [37] and Feng et al. [49] who used quota sampling, the remaining 15 articles used convenience sampling to recruit the target samples. Additionally, only one article from a developing country was identified [48]. One reason for this may be the higher incidence of eye cancer in Europe and the US, which means that more people are affected; moreover, these countries have sufficient experience and available tools compared to other countries. However, because of the lack of advanced medical conditions in developing countries, the long-term survival rate of patients with eye cancer remains unchanged, with shorter survival times [49]. All the above showed that the samples may not well represent the overall QOL worldwide. Sampling variability may be related to the location of the participants and sampling error. Therefore, it would be beneficial to seek out and include populations from different regions in future studies, which may clarify the existing inconsistencies.

As for the Publication bias, first, among the included articles, 5 of 17 were published in two journals, including three in Ocul Oncol Pathol [5,16.41] and two in Acta Ophthalmol [44, 45]. second, as for the authors, Hope-stone, Damato, Brown et al. were in the same team, and published five articles in 2016, 2018, 2019, and 2021 with patients treated in different periods and measured with different questionnaires at different time [28, 30, 40,41,42]. Similarly, Gollrad et al. [46, 48] published two articles in 2021 and 2022 with different target population at different periods. Klingenstein et al. [37,53] also published two articles in 2013 and 2015 using the same questionnaire but measured at different time points, but part of the data was obtained by extending the follow-up time based on the data in 2013. Third, positive results and large sample results are easier to publish. Among the studies included in this review, 11 (i.e., positive results) considered QOL to be worse than the general population, accounting for 64.71%. Luckly, we initially searched the Clinicaltrials.gov which may include some grey literature to ensure comprehensiveness. Forth, only seven research articles (41.18%) [17, 30, 33, 44,45,46,47] mentioned fund support but with no conflict of interest, which was also an indicator to evaluate publication bias. Fifth, we excluded non-English articles due to language barriers, which may lead to language bias. All of the above have reduced publication bias to a certain extent.

The results also lacked comparative and normative data for healthy controls. Without a standardised questionnaire, the results varied significantly. Furthermore, the included studies used a variety of methods, with different follow-up periods, sample sizes, and demographics, which resulted in heterogeneity, all of which make it difficult to generalise the findings from one study to larger patient populations. Compared to patients with UM, RB patients experience relatively fewer functional problems, and less anxiety and depression, which may be because RB survivors survive longer than patients with UM, and RB occurs more frequently in children; when they grow up, they usually cherish life more with high resilience than healthy people [16, 49].

Limitations and next-step recommendations

This systematic review had some limitations. First, there were no randomised controlled trials in the included studies and only patients with RB and UM were included, which may not fully represent eye cancer patients. Second, because of the heterogeneity of the treatments, assessment tools and different follow-up times in each study, it was difficult to effectively compare the data included in this review. Therefore, it is necessary to develop a standardised QOL assessment tool for patients with eye cancer. Further, it is recommended that more attention be paid to developing a specific standard assessment tool, focusing on the QOL of eye cancer survivors in low- or middle-income countries and taking measures to improve their QOL.

Conclusion

Through this review, it is found that there are differences in the current QOL of patients with eye cancer, with anxiety and depression occurring more frequently. However, it is difficult to assess QOL accurately because of the differences in treatment, assessment tools, and follow-up times. Therefore, it is necessary to develop a standardised QOL assessment tool and a follow-up protocol. It is worth noting that more attention should be paid to developing countries, whose mortality rates are high, to improve their QOL.