Since the mid 1990s, an emergent body of literature has demonstrated an association between chemotherapy treatment for breast cancer and changes in cognitive functioning, rendering cognition an important cancer ‘survivorship’ issue. Results from early cross-sectional investigations indicated that the percentage of women experiencing cognitive impairment following chemotherapy ranged from 16% to 50% (Vardy & Tannock, 2007). Prospective, longitudinal designs incorporating a pretreatment assessment have generally reported lower overall incidences of cognitive impairment than cross-sectional approaches. Indeed, cognitive declines following chemotherapy treatment appeared to be relatively subtle, experienced by only a subset of women undergoing treatment, specific to a narrow range of cognitive domains, and most prevalent during the acute treatment phase. In general, however, these studies were limited by relatively small sample sizes, in the length of the follow-up period, and in the statistical techniques employed to analyse their data, often relying on analyses that used mean performance scores to make comparisons. However, given that only a subset of women appear to demonstrate cognitive decline after chemotherapy, it is possible that the performance of the unaffected majority obscured that of the clinically relevant minority.
Thus, the initial aims of this thesis were twofold: to employ a longitudinal modelling approach to investigate trajectories of cognitive performance across a two year period encompassing treatment and long-term recovery; and to examine a wide range of potential risk factors for, and predictors of, chemotherapy-induced cognitive change. A prospective, longitudinal design was employed to assess 123 women undergoing chemotherapy treatment for breast cancer at four timepoints: Time 1 - after surgery but before commencing chemotherapy treatment; Time 2 - one month following completion of chemotherapy treatment; Time 3 - six months following completion of chemotherapy treatment; and Time 4 - 18 months following completion of chemotherapy treatment. At each assessment timepoint, the women completed a range of neuropsychological tests encompassing several cognitive domains, as well as self-report measures of psychological well-being and perceived cognitive functioning.
In Chapter 3, the novel application of a longitudinal modelling approach was reported. This was successful in identifying subgroups of women who appeared to be differentially affected by chemotherapy treatment. Overall, for most of the cognitive domains tested, the performance trajectories of the women remained stable or indeed improved over time; however, the low trajectory subgroup was often characterised by no improvement in contrast to the other subgroups. Performance on verbal memory tasks declined significantly across the treatment period for all trajectory subgroups; however, while the high and mid trajectory subgroup showed an improvement during recovery which appeared to take them back to baseline levels, the low subgroup did not. This was an important finding because it served to identify a subgroup of women experiencing prolonged verbal memory impairment, with no evidence of recovery even 18 months after the completion of chemotherapy.
Chapter 4 reported the results of a series of logistic regression analyses to explore whether certain demographic, treatment/health/disease, or psychological well-being characteristics served as predictors of cognitive functioning trajectories following chemotherapy treatment. Though no predictors emerged consistently, being older and having a lower IQ score increased the likelihood of having a low cognitive performance trajectory on several measures, while having a later stage of cancer acted as a predictor for a low trajectory on the memory measures exclusively. It was considered possible that stage of cancer may influence cognitive performance through an indirect treatment-related mechanism. Alternatively, the diagnosis itself of a later stage cancer may influence an individual’s perception of the severity of the disease, and perhaps the severity of the side-effects of treatment.
The role of individual self-perception was explored in Chapter 5 in a study that aimed to investigate the course of perceived or subjective cognitive functioning following chemotherapy, as well as the relationships between subjective and objective measures of cognition. The hypothesis that women would experience a decline in subjective cognitive functioning across the treatment period, followed by an improvement over the recovery period, was supported. However, the results indicated that even 18 months after the end of chemotherapy, participants’ perceptions of their cognitive functioning had not returned to the level they experienced prior to the start of treatment. In line with previous research, subjective cognitive functioning did not correlate with any neuropsychological measure of objective cognition. However, lower levels of subjective cognitive functioning were significantly associated with higher levels of psychological distress. This suggested that the mechanisms underlying objective and subjective cognitive change following chemotherapy in breast cancer patients may be quite different.
Together, the findings of this thesis have important clinical implications for identification of individuals ‘at risk’ of experiencing cognitive change following chemotherapy, as well as for the design of interventions to address this important breast cancer survivorship issue.