Background

Cancer-related cognitive impairment (CRCI) and its impact on individuals and families are gaining attention in cancer survivorship. In prostate cancer, one of the most common malignancies worldwide reported cognitive side-effects from hormonal treatments are not uncommon. Prevalence ranges between 10 and 69% of prostate cancer survivors (PCS) [1]. Hormonal treatments for prostate cancer involve suppressing production of androgens (e.g. testosterone, estradiol) to reduce tumour growth. This can be achieved by decreasing androgen production (e.g. luteinizing hormone-releasing hormone agonists, gonadotropin-releasing hormone agonists, orchiectomy, CYP17 inhibitors), and/or blocking androgen in the body (e.g. anti-androgens, androgen receptor antagonists) [2]. Hormonal treatments are often combined with other treatment modalities (radiation therapy, surgery, chemotherapy) and can be administered continuously or intermittently [2].

Increasingly research supports the role of sex hormones, such as androgens, in cognitive functioning [3]. Androgens and their metabolites have been demonstrated to have neuroprotective effects in maintaining cognitive functions [4]. Cognitive changes following hormonal treatments for prostate cancer often occur in the context of an ageing population already at risk of developing cognitive impairments (mean age at diagnosis is around 65 years), where a myriad of age- and treatment-related factors (e.g. biological, psychological, socio-environmental, cancer and accumulated toxicities from treatment) likely contribute to cognitive changes [5]. Other factors may include cancer stage, comorbid health conditions or impacts of hot flashes, fatigue, sleep disturbances and psychological morbidity (e.g. depression, anxiety, post-traumatic stress) [6].

CRCI is often mild to moderate in nature and can be transient or more persistent [7]. The complex interplay of aforementioned factors underlying CRCI in PCS likely contributes to reported brain fog and disruptions to attentional processes leading to declines in other cognitive domains such as memory and higher-order executive functions [8,9,10,11,12,13,14]. These cognitive difficulties may significantly impact quality of life, interpersonal, psychological, role and occupational functioning, as well as autonomy and safety, especially in older individuals [15]. Moreover, PCS on hormonal treatments often report increased emotional lability, irritability and anger [16]. Challenges with emotional and/or behavioural regulation likely impact relationship dynamics. However, little is known about any possible interaction between CRCI and relationship dynamics in PCS.

There is limited research exploring PCS experiences of CRCI following hormonal treatment and the impact of these changes on their lives and their families. This study aims to qualitatively explore the nature and impact of cognitive changes following hormonal treatments in PCS from the perspective of PCS and their partners.

Methods

Design

We used an inductive qualitative approach guided by principles of Interpretative Phenomenological Analysis (IPA) [17] applied with the “Framework Method” [18]. IPA is a participant-oriented approach recognising the double hermeneutical process. Since cognitive difficulties (e.g. word-finding problems, attentional or memory lapses) may be exhibited during interviews, the IPA approach facilitated the analysis of these behaviours. The Framework Method allowed for the integration of an inductive and deductive approach using a systematic method of organising and categorising data into a matrix. The Consolidated Criteria for Reporting Qualitative Research (COREQ) Checklist was used to guide the reporting of this study [19].

Recruitment

Participants were recruited from a larger cross-sectional survey investigating cognitive functioning in PCS. This study was registered with Australian New Zealand Clinical Trials Registry on the 25th November 2021 (registration number ACTRN12621001608853). English-speaking PCS, aged 18 years and older, on hormonal treatments (e.g. androgen deprivation therapy) or on “watchful waiting”/ “active surveillance” were recruited from Australian clinics, healthcare professionals, email invitations, research registers (i.e., HealthMatch), social media (i.e., Facebook, Twitter), community groups and by word of mouth between January 21, 2021, and October 15, 2023. Participants receiving hormonal treatments, consented online to be contacted about participating in an interview and endorsed problems with cognition, were invited via phone call or email to participate in an interview. The criteria for identifying PCS with CRCI involved either endorsing “yes” to a list of treatment side-effects including “changes in thinking (e.g. memory, attention)”, or scoring ≥ 1 standard deviation below the mean of non-cancer controls on the Perceived Cognitive Impairments subscale from the Functional Assessment of Cancer Therapy—Cognitive Function (FACT-Cog) and/or the cognitive functioning scale from the European Organisation for Research and Treatment of Cancer Quality of Life Questionnaire (EORTC QLQ-C30). This conservative approach included PCS who might have milder cognitive difficulties compared to their similar age peers without prostate cancer. Partners were recruited through PCS and provided separate consent. Ethical approval was obtained from Macquarie University Human Research Ethics Committee (reference number: 52020611919011), in accordance with the NHMRC National Statement on Ethical Conduct in Human Research (Commonwealth of Australia, 2007- Updated May 2015) and the Declaration of Helsinki.

Procedures

Semi-structured interviews were conducted by Author 1 in the participants’ preferred format. Participants selected interview via telephone or videoconferencing, which were audio-recorded. Interviews were semi-structured allowing flexibility in response-driven inquiry. Topics discussed included onset of cognitive changes, nature and impact of cognitive difficulties, coping and neuropsychological interventions (see Supplementary Materials). PCS reported on their own experiences, whilst partners of PCS provided their perspective on the PCS’s cognition.

Consent was confirmed verbally and recorded in the interview. PCS and their partners were interviewed separately, privately and only once, except for one couple who requested a joint interview. Notes and initial reflections were recorded both during and after each interview. As number of individuals suitable for interview was limited by recruitment from a cross-sectional survey and eligibility constraints, the concept of information power was used to determine when sufficient richness of data had been obtained and interviews stopped [20]. All interviews were transcribed by Author 1 and research interns. Transcripts and findings were not returned to participants for review.

Analysis

Interview transcripts were coded with themes identified using both IPA guidelines and the Framework Method. Firstly, brief descriptive summaries of participants’ experiences of cognitive changes following prostate cancer treatment were produced for each transcript to allow the researchers to be “participant-observers”. Open coding was employed for each transcript, identifying concepts arising, making comments on language, repetition, contradictions, followed by phenomenological coding (i.e., line by line analysis). Tangential speech was coded as responses that were irrelevant to the questions asked or general topic of conversation, and/or identified by participants themselves as off-topic. After multiple readings of transcripts, concepts arising from initial steps were clustered into superordinate themes and subthemes. Research interns met with Author 1 to discuss the themes, identifying quotes to characterise and support themes. An experienced qualitative researcher (Author 2) oversaw the analysis. Data were compared across individual participants to develop a working then final analytical framework. Lastly, transcripts were recoded and data were charted into the framework matrix for interpretation. Any themes lacking rich evidence to add value to the emerging framework were eliminated. Qualitative data were managed and organised using N-Vivo 20 (QSR International, 2021). Methodological rigour was maintained through acknowledgement of reflexivity, team discussion and debriefing and iterative changes to the interview guide.

Results

The sample consisted of 11 participants residing in Australia: six PCS and five partners (all female). An additional two partners did not wish to participate in the interviews due to time constraints or the topics being “too difficult to talk about”. Mean interview duration was 52 min, ranging from 20 to 71 min. The median age of our sample was 72 years (range 61 to 79), full demographic details and cancer characteristics of PCS are outlined in Table 1. Five themes were identified from the analysis, which are elaborated below (see coding tree in Supplementary Materials).

Table 1 Cancer characteristics of men with prostate cancer on hormonal treatments
Full size table

Theme 1: Diverse nature of cognitive difficulties and its challenges

Table 2 presents example quotations of the diverse nature of cognitive difficulties and their challenges. PCS reported a range of cognitive difficulties, verified by their partners, including difficulties with attention, memory, word-finding, organization, slowed processing speed and general cognitive decline. Most PCS described feeling “not as sharp” or more “foggy”, often associated with fatigue, slowed speed of processing, attentional problems and mistake-making. These interrelated difficulties were more pronounced during complex tasks requiring increased attentional processing and resulted in increased time and effort to complete familiar tasks.

Table 2 Quotes to illustrate the diverse nature of cognitive difficulties and its challenges
Full size table

Difficulties with short-term memory and “forgetfulness” were prominent concerns for participants. This was characterised by inefficient or unreliable encoding of information. Some men employed external strategies to aid their memory or often relied on their partners to prompt and remind them. Moreover, repetition was necessary to compensate for deficits in information processing and poor retention.

Word-finding problems were reported by most PCS and were observed during interviews. This often disrupted the flow of conversation and impacted expressive meaning.

Few PCS reported higher-order executive function problems (e.g. problem-solving, decision-making, planning), except for increased disorganisation. Qualitatively, some PCS demonstrated tangential speech during the interview, on several occasions forgetting the question asked.

Confidence and concerns about driving were also raised by both PCS and their partners, with partners assuming increased responsibility in driving and/or helping the PCS navigate.

Theme 2: Uncertainty about aetiology of cognitive difficulties

Quotations supporting the following responses regarding the aetiology of cognitive difficulties are presented in Table 3. Since many PCS received a combination of treatments, most expressed uncertainty as to whether their cognitive changes were solely due to hormonal therapy or compounded treatments and their side-effects (e.g. fatigue, hot flashes, sleep problems). Hormonal treatments were also perceived as exacerbating pre-existing conditions. PCS acknowledged difficulty distinguishing between “normal” age-related processes and treatment-related cognitive decline, as well as the potential of treatment exacerbating the normal ageing process. Furthermore, treatments may have increased awareness of cognitive changes.

Table 3 PCS beliefs about the aetiology of cognitive difficulties
Full size table

Theme 3: Shifts in self-perception, interpersonal dynamics and increased emotionality

Table 4 presents example quotations supporting the following expositions of this theme. Difficulties with cognitive functioning were reported to lead to shifts in self-perception. A loss of confidence, and poorer quality of life and daily functioning were reported by PCS. Decreased confidence, fatigue along with cognitive side-effects were reported to lead to increased social withdrawal and potentially leaning more into pre-existing traits (e.g. introversion). For instance, PCS indicated perceptions of accelerated cognitive decline led to decreased engagement in professional settings and at family gatherings.

Table 4 Quotes to illustrate shifts in self-perception, increased emotionality and interpersonal dynamics
Full size table

The cancer experience and cognitive difficulties were reported to lead to shifts in relationship dynamics, especially interpersonal interactions and responsibilities between PCS and family members, which were reported by all partners interviewed. The sentiment of “I am doing more things, I’m doing a lot more things” (PID01-P) was present in many partner accounts, “doing more” included more driving, providing reminders, and household tasks. Such changes sometimes shifted “power” dynamics and led to greater appreciation for their partners. Moreover, several participants reported increased irritability and emotional lability, which impacted interpersonal dynamics.

Theme 4: Coping and management

Table 5 presents quotations supporting the following range of different coping and management strategies employed by PCS. No PCS sought professional help for their cognitive difficulties, instead cognitive difficulties were managed by engaging in cognitively stimulating activities and compensatory strategies. Most PCS were proactive in "keeping busy” and keeping their brain “active” with a range of cognitively stimulating activities, including establishing a men’s community organisation called “Men’s Shed”, being involved in building management, house and garden maintenance, managing paperwork for a band, writing sermons, engaging in casual work, socialising, and engaging in a “brain training” programme. Using external aids, such as lists and a diary, were the primary compensatory strategy used to aid memory, planning and organisation. However, consistent implementation of external aids was necessary to support cognition. Interestingly, some reported changes in attitude, becoming more relaxed, as a means of coping with cognitive changes.

Table 5 Range of coping and management strategies employed by PCS
Full size table

PCS also employed a range of strategies to cope with the impact of cancer more generally, including antidepressant medication and engaging in health promoting behaviours. Support from allied health professionals was highly valued in providing holistic cancer care such as prostate cancer nurses. Cancer centres were helpful in providing holistic care. Information-seeking was another coping behaviour employed by both men and their partners in managing the impact of cancer. Participants also reported the benefits of social support from family, friends and community groups and pets. Keeping positive was another coping mechanism reported involving both mental and behavioural processes.

Theme 5: Opinions on a neuropsychological group intervention

Table 6 presents quotations supporting the following opinions of a neuropsychological intervention from PCS and their partners. Whilst PCS received support for physical side-effects of treatment, there was little support for the cognitive side-effects. All participants affirmed the potential benefits of a group neuropsychological intervention, if cognitive rehabilitation/remediation were possible.

Table 6 Opinions on a neuropsychological intervention
Full size table

Participants shared a range of ideas potentially important for development of a neuropsychological intervention. These included the need for a “roadmap” to encourage a positive direction and contribution from PCS going through similar experiences. Information should be personalised and patient-centred. In addition, strategies to aid cognitive functioning should tailored to the concerns noted by PCS and family members.

Qualities of a good intervention facilitator included being a good listener, empathetic and having group work skills to create a positive, safe, welcoming space. However, there was some apprehension towards having a “large” group, with a smaller group or “buddy” system being more desirable.

Discussion

Our results highlight the nuanced and diverse range of perceived cognitive difficulties experienced by PCS undergoing hormonal treatments. They indicated impacts on self-perception, daily functioning and interpersonal dynamics. There was general interest and specific feedback regarding the potential for a neuropsychological intervention, suggesting PCS and their partners would engage with such an intervention.

The diverse nature of cognitive difficulties and their associated challenges reflect previous research. Decline in attention, processing speed, verbal/language skills, memory, visuospatial and visuomotor skills and executive functioning have been identified in PCS on objective neuropsychological tests following hormonal treatments [8,9,10,11,12,13]. Although findings have been inconsistent [21, 22], this suggests a subset of PCS may be more vulnerable to experiencing cognitive changes following hormonal treatments. Moreover, as our qualitative data demonstrate, cognitive decline may remain undetectable to others despite self-reported challenges of PCS in participating in daily activities (e.g. driving, social engagement) and reduced productivity [23]. Reported difficulties on more cognitively demanding tasks add to earlier findings indicating hormonal treatments likely disrupt complex information processing (i.e., using higher-order executive processes) rather than specific cognitive domains [14]. Such deficits may manifest more evidently in ecologically valid contexts (e.g. work environment) over standard neuropsychological testing conditions. This is important given cognitive complaints may increase the likelihood of early retirement in older cancer survivors [24]. Therefore, addressing cognitive complaints may play an important role in effective return to work and optimising productivity, impacting both personal wellbeing (i.e., quality of life, mental health and financial stability) and societal economic health more generally [25].

Our respondents were unclear regarding the aetiology of cognitive changes following hormonal treatment, as these changes were co-occurring with other potential causal factors including other cancer treatments, ageing, co-morbid illnesses and psychological factors (e.g. depression, anxiety, stress). For instance, ageing-related decreases in testosterone levels have been associated with decline in memory functioning, processing speed, visuospatial skills and visuomotor abilities in older men [26], which are consistent with neuroanatomical areas containing androgen receptor expression (hippocampus, amygdala, etc.) [27]. Development of CRCI is likely multifactorial [28] associated with the cancer itself [29], cancer treatments, exposure to hormonal treatments [30], genetic susceptibilities and other medical comorbidities [31]. Cancer and its treatments have also been associated with chronic neuroinflammation, DNA-damage, endothelial dysfunction, all increasing risk of brain ageing [32]. Moreover, our data highlighted the perceptions of accelerated brain ageing following treatment, which support preliminary neuroimaging data [33]. Additionally, we explored how clusters of co-occurring, interrelated symptoms (e.g., hot flashes, sleep problems, insomnia, fatigue and depression) may lead to cognitive deficits. Causal relationships, however, are currently speculative, requiring further investigation. Furthermore, older individuals are likely more sensitive to declines in androgen levels, treatment-related toxicities and disturbances in physical and psychological functioning, increasing the risk of CRCI [5].

PCS in our study reported on the impact of treatment side-effects on self-perception and emotionality that in turn affected relationship dynamics. Novel insights into the perceived impact of cognitive or intellectual functioning on self-image in PCS were identified in our sample of highly educated men, who may have been more sensitive to decrements in cognitive performance. Our data suggest such decrements occurred along with decreased self-confidence, which accompanied deficits in occupational functioning and social withdrawal, leading to shifts in interpersonal relationships. Whilst couples commonly report role shifts after a cancer diagnosis, changes in role functioning (e.g. employment, independence, personal responsibilities) may challenge core beliefs (e.g. self-image and perceptions of masculinity) [34]. The nature of relational shifts and extent to which they are a consequence of CRCI requires further investigation, especially as partners of PCS are likely part of an ageing population experiencing normative cognitive decline and potentially having challenges of their own. Moreover, whilst treatment side-effects like urinary incontinence have been found to lead to social isolation in PCS [35], our results provide a unique perspective of the impact of cognitive side-effects, consistent with research more broadly on cognitive decline and social withdrawal [36]. Our data also highlighted how changes in interpersonal relationships may be further complicated by increased emotional lability following treatment. Thus, neglecting to address the impact of hormonal treatments on partners and the dyadic relationship may diminish the physical and psychological wellbeing of PCS [37].

Despite all participants acknowledging the potential benefits of a neuropsychological intervention, none had sought professional help for their cognitive difficulties. This was partly due to a lack of information provided by treating professionals and limited availability of interventions, consistent with reviews of the research [15]. Lower rates of help-seeking for PC-related concerns have also been reported in the literature with around 40% of PCS not seeking support from health professionals or support services [38]. Moreover, 82% of men reported unmet supportive care needs with 47% reporting unmet psychological needs [38]. Factors associated with lower help-seeking included older age, low cancer knowledge, elevated depressive symptoms, embarrassment, fear and adherence to masculine norms (e.g. self-reliance, stoicism, appearance of control) [39, 40]. Addressing barriers to help-seeking is pivotal, given the apparent conflict between perceived supportive care needs and help-seeking behaviours in PCS. Some solutions presented by Cowan et al. (2023) include encouraging healthcare providers to facilitate conversations around CRCI with PCS, identifying people at risk of developing CRCI, conducting assessment of CRCI and providing therapeutic support and management of these symptoms [15].

Regarding the management of CRCI, we found PCS applied range of compensatory strategies (e.g. note-taking), but their level of success varied. Enhancing the effectiveness of these strategies may involve neuropsychological interventions like cognitive rehabilitation. Whilst there is growing evidence supporting the benefits of cognitive rehabilitation in cancer populations [41], its effectiveness in PCS has not been determined. It is possible broader emotional or psychological support will also assist in coping with CRCI. Additionally, prostate cancer nurses can play a crucial role in connecting individuals to these supports and may also help overcome reluctance to help-seeking [42].

Our study has several limitations. Whilst the use of qualitative methodology allows for in-depth exploration and understanding of participants’ experience, it is characterised by self-selection bias. The sample primarily consisted of retired, older, Caucasian men with advanced prostate cancer, potentially increasing their susceptibility to CRCI due to treatments effects, associated with its type and duration, and tumour metastases [28]. Moreover, younger PCS still in the workforce may have different experiences and needs compared to current participants, highlighting the need for further research in this area. There were also no participants treated only with hormonal therapies, which meant we could not tease out the cognitive impacts of hormone therapies alone. Though perceived cognitive functioning often poorly correlates with objective measures and is more closely associated with psychological factors [43, 44], it may better reflect the day-to-day experiences of cancer survivors compared to standardized neuropsychological measures and tend to be a stronger predictor of quality of life and functional outcomes [45].

PCS may experience a diverse range of cognitive difficulties. Their aetiology is likely multifactorial. Cognitive difficulties may negatively impact self-perception, daily functioning and interpersonal dynamics. Whilst neuropsychological interventions addressing the cognitive, psychological and behavioural concerns may be of benefit, it is critical to incorporate current approaches to working therapeutically with older males to increase engagement of PCS with any intervention.