This study analyzed disease-specific aspects of HRQoL in a large population-based cohort of long-term BC survivors (5–15 years post-diagnosis). In general, we found that age, education, and, to some lesser extent, treatment- and disease-related variables were associated with disease-specific HRQoL. BC survivors with lower educational levels reported more symptoms than those with higher education, especially at younger ages. It has been reported before that socio-economic deprivation of BC survivors is associated with lower quality of life, which might be explained by lower uptake of follow-up care and more stress due to lower socio-economic status (Dialla et al. 2015).
BC survivors reported poorer body image after mastectomy, in which breast reconstruction only showed a beneficial effect for body image in the age group 30–49 years, but not for women aged 50 years and above. This is in line with a systematic review finding that breast reconstruction is more relevant for body image in younger BC survivors (Paterson et al. 2016). According to another systematic review, previous studies on the effect of breast reconstruction in general found discrepant results for different aspects of body image (Fang et al. 2013). Previous studies also found breast reconstruction to be less beneficial for BC survivors’ body image than expected (Rowland et al. 2000; Janz et al. 2005). The process of breast reconstruction usually takes more time and might be more stressful than mastectomy alone (Rowland et al. 2000), and coping with the cosmetic result could be more difficult than coping with mastectomy (Janz et al. 2005). Some BC survivors might be disappointed with the outcome or with the fact that breast-conserving treatment was not possible (Janz et al. 2005; Rowland et al. 2000).
Despite the long time since diagnosis, BC survivors in age groups 60 years and above reported more breast and arm symptoms after lymph node dissection and radiotherapy. However, results were mixed and in some age and treatment subgroups there was no difference in arm and breast symptoms according to treatment. A previous prospective study showed that an effect of axilla surgery on arm problems was dependent on the number of harvested lymph nodes (Engel et al. 2003), and in our study, this information was not available. Overall, breast and arm symptoms decreased with increasing age and were more a problem of those BC survivors still at working age, which is in line with previous studies (Engel et al. 2003; Macdonald et al. 2005).
Body image and future perspective were positively associated with age, and upset by hair loss was less of a problem in older survivors. As they age, BC survivors might shift values, have a more relaxed view on themselves and the future, and are more at peace with their physical appearance. A systematic review showed that body image was lower in younger BC survivors, but without a difference to same-aged controls (Davis et al. 2020).
The BC survivors in this study had long completed their therapy and we did not find any long-term associations between chemo- or endocrine therapy and systemic therapy side effects. Further, only a few BC survivors reported (permanent) hair loss, leading to a small sample size who answered the item “upset by hair loss”. Nevertheless, we found that younger survivors were more likely to be upset by hair loss. However, no statistically significant associations of being upset by hair loss were found according to endocrine therapy, and associations with chemotherapy were only visible in the overall group, but not in age-stratified analyses. When receiving chemotherapy, affected women might know that hair loss is mostly temporary, so the differences are not reflected in upset about hair loss. Hair loss might also be a reason to terminate endocrine therapy early (Karatas et al. 2016), which was not controlled for in our study. We also did not explicitly assess hair thinning or other qualitative changes of (regrown) hair. According to a Japanese study, 59% of BC survivors 4–5 years post-diagnosis reported a scalp hair recovery of > 80%, 35% reported only 40–70% recovery, and in general, the hair was often reported to regrow thinner, curlier or greyer (Watanabe et al. 2019).
In our study, sexual functioning, which in the questionnaire relates to sexual interest and the frequency of being sexually active, was lower at older age. The decrease in sexual activity with age, irrespective of treatment, is in line with previous studies in long-term BC survivors (Soldera et al. 2018; Ganz et al. 2002). However, sexually active women in all age groups reported comparable levels of sexual enjoyment.
There was no clear association between tumor stage at diagnosis and disease-specific HRQoL. We only found one statistically significant difference in the age group 70–79 years, in which BC survivors with stage IV reported a better future perspective than stage I and II. This specific subgroup of elderly survivors with initially poorer prognosis are still alive longer than expected based on their age and stage (mean time since diagnosis in this group: 9.0 years), which could have resulted in more hope and optimism. However, given that the pattern is solely found in one age group and the number of survivors with stage IV is small, it is unlikely that the result reflects a systematic effect.
BC survivors below 80 years with active disease reported a worse future perspective than disease-free BC survivors of the same age. Cancer recurrence is often associated with hopelessness, psychological stress (Brothers and Andersen 2009), and lower quality of life (Arndt et al. 2005). Active disease was also associated with more side effects in certain age groups, which might be due to additional or more intensive or burdensome treatments. Although the analyses were adjusted for treatments (including primary treatment and treatment of recurrence), the intensity of those treatments, like chemotherapy doses or radiation circles were not assessed and thus could not be considered for adjustment. We also did not consider the year when the treatments were undergone, as this information was reported only by a subset of the respondents.
Some further limitations have to be considered when interpreting the study results. There is a risk of healthy survivor bias, as BC survivors who were of higher age and had more health problems might have been less likely to participate in the study. This would result in an underestimation of remaining health problems. We did not compare functioning and symptoms to normative data or a control group as the BR23 is not applicable in women without breast cancer. Further, participants were recruited 5 and more years after diagnosis and the study was cross-sectional. As such, a potential underassessment of the risk for late and long-term sequelae due to a prevalence-incidence bias (Neyman bias (Hill et al. 2003)) has to be considered. Treatments were self-reported by yes-/no-questions and might be biased due to memory effects, although we assume that most participants remembered burdensome treatments like operations or chemotherapy. The timing and intensity of treatments were not considered in this study. Further, some treatments were reported by a vast majority of the sample (e.g. lymph node dissection) which led to very unequal group sizes for comparison. Some treatments and guidelines might have changed since the time when the study cohort was diagnosed, e.g. more modern, less invasive or lower-dosed treatments, newer therapeutic agents or treatment regimens, or a lower rate of lymph node dissection at earlier stages (Fisher et al. 2020). In recent years, shared decision-making has also gained more popularity. Newly diagnosed BC patients might have more possibilities to actively participate and make an informed treatment decision, compared to the BC survivors in our sample, which might result in a better sense of control over potential late-effects.
When analyzing active disease, we did not control for the date of recurrence, metastasis or second cancer due to the overall small number of such events. Differences in functioning and symptoms according to age reflect the cross-sectional comparison of age cohorts rather than the longitudinal development of individuals. Some important information such as cancer stage was not fully reported to the cancer registries in the period when study participants were first diagnosed (7.7% missings). To overcome this, we replaced missing data using multiple imputation and ran sensitivity analyses which showed that the imputed results were similar to those from non-imputed data. Another limitation of the study is that we restricted the subgroup analyses on disease-specific problems to age, education, clinical and treatment-related variables. Future studies might focus more on sociodemographic and psychosocial factors like job situation or social support to identify special support needs and, if needed, starting points for intervention.
Strengths of the study include the large sample size that was recruited in a multi-regional, population-based study. The large sample size allowed us to stratify groups for more in-depth analyses and adjust the results by various clinical and sociodemographic variables, increasing the validity of the results for different subgroups of long-term BC survivors.
In a previous study, we have shown that the disease-free subsample of BC survivors in our study reported a HRQoL that was comparable to that of same-aged population controls (Doege et al. 2019). Similarly, it is heartening that in the current study, BC survivors reported, in general, no potential long-term side effects of self-reported chemotherapy and endocrine therapy. However, for BC survivors in certain age groups, some disease-specific issues were still of concern even long after diagnosis. For example, younger survivors reported a worse body image after mastectomy and more arm and breast symptoms after lymph node dissection and radiotherapy. Therefore, it is important for clinicians and for BC survivors to know that some symptoms and functioning deficits can persist over years and that needs of long-term BC survivors differ according to age, education, treatment, and disease status. BC patients diagnosed in recent years might profit from advances in shared decision-making, enabling them to find their personal balance between risks and benefits of the treatments they undergo, e.g. regarding breast reconstruction. Patients with lower educational levels and potentially lower health literacy might profit from clear and simple explanations regarding treatment options and potential side effects to gain more sense of control. BC survivors should be encouraged to bring potential ongoing physical and psychological problems and concerns to the attention of health care providers. Further research should evaluate if and to what extent a standardized screening of patient-reported outcomes at different stages of the survivorship trajectory might help to identify specific needs and to support vulnerable BC survivors to improve their coping process.
留言 (0)