Brain, Behavior, and Immunity: Frontal gray matter reduction after breast cancer chemotherapy and association with executive symptoms: A replication and extension study

Cognitive changes related to cancer and its treatment have been intensely studied, and neuroimaging has begun to demonstrate brain correlates. In the first prospective longitudinal neuroimaging study of breast cancer (BC) patients we recently reported decreased gray matter density one month after chemotherapy completion, particularly in frontal regions. These findings helped confirm a neural basis for previously reported cognitive symptoms, which most commonly involve executive and memory processes in which the frontal lobes are a critical component of underlying neural circuitry. Here we present data from an independent, larger, more demographically diverse cohort that is more generalizable to the BC popula­ tion. BC patients treated with (N = 27) and without (N = 28) chemotherapy and matched healthy controls (N = 24) were scanned at baseline (prior to systemic treatment) and one month following chemotherapy completion (or yoked intervals for non-chemotherapy and control groups) and APOE-genotyped. Voxel­ based morphometry (VBM) showed decreased frontal gray matter density after chemotherapy, as observed in the prior cohort, which was accompanied by self-reported difficulties in executive function­ ing. Gray matter and executive symptom changes were not related to APOE e4 status, though a somewhat greater percentage of BC patients who received chemotherapy were e4 allele carriers than patients not treated with chemotherapy or healthy controls. These findings provide confirmatory evidence of frontal morphometric changes that may be a pathophysiological basis for cancer and treatment-related cognitive dysfunction. Further research into individual risk factors for such changes will be critical for development of treatment and prevention strategies. © 2012 Elsevier Inc. All rights reserved. Cognitive changes related to breast cancer and its treatment have been an area of increasing study, with numerous reports demonstrating cognitive impairment in patients relative to con­ trols. These changes have been differentially attributed to chemo­ therapy, radiation, and anti-estrogen treatment (Agrawal et al., 2010; Ahles et al., 2010; Collins et al., 2009; Jim et al., 2009; Ques­ nel et al., 2009), and have been reported most prominently in exec­ utive functions (e.g., working memory) and processing speed, cognitive processes largely subserved by frontally mediated brain systems (impairment in other cognitive domains has also been noted; for review and meta-analysis see (Anderson-Hanley et al., 2003; Correa and Ahles, 2008; Stewart et al., 2006)). A higher than expected incidence of impaired cognitive performance has also been found in patients prior to systemic treatment (Ahles et al., ⇑ Corresponding authors. Address: Center for Neuroimaging, Department of Radiology and Imaging Sciences, Indiana University School of Medicine, 950 W. Walnut St., R2 E124, Indianapolis, IN 46202, United States. Tel.: +1 317 274 2067; fax: +1 317 274 1067. E-mail addresses: [email protected] (B.C. McDonald), [email protected] (A.J. Saykin). 2008; Wagner et al., 2006; Wefel et al., 2004), suggesting that host factors and/or the cancer disease process itself may play a role. This prior work demonstrates the continued need for further investiga­ tion of the effects of cancer treatment and the disease process on cognition in vulnerable individuals (McDonald and Saykin, 2011; Vardy et al., 2008). The neural mechanisms underlying these cognitive changes have likewise been the subject of increasing investigation. Several cross-sectional, retrospective structural MRI studies have utilized voxel-based morphometry (VBM) to assess gray matter changes after breast cancer treatment quantitatively, in an automated, unbiased manner (de Ruiter et al., in press; Hakamata et al., 2007; Inagaki et al., 2007; McDonald et al., 2008; Saykin et al., 2003; Yoshikawa et al., 2006). Those studies comparing gray mat­ ter between patients who did and did not receive chemotherapy have demonstrated residual gray matter deficits in the chemother­ apy-treated group, even several years after treatment completion (de Ruiter et al., in press; Inagaki et al., 2007; McDonald et al., 2008; Saykin et al., 2003). We recently reported the first prospec­ tive VBM study examining such gray matter changes relative to pre-treatment baseline (McDonald et al., 2010). We predicted that 0889-1591/$ see front matter © 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.bbi.2012.05.007 S118 B.C. McDonald et al. / Brain, Behavior, and Immunity 30 (2013) S117–S125 these changes would be detectable in the short-term but would recover at least partially over time, given prior cognitive studies suggesting longitudinal improvement in brain function after che­ motherapy (Ahles et al., 2010; Collins et al., 2009; Jansen et al., 2011; Jenkins et al., 2006; Schagen et al., 2002). Findings were con­ sistent with study hypotheses, demonstrating reduced gray matter in chemotherapy-treated patients one month after chemotherapy completion in bilateral frontal, medial temporal, and cerebellar re­ gions. One year later gray matter density had returned to baseline levels in some regions, though not all. No between-group differ­ ences were found at baseline, and changes were not seen in pa­ tients who did not receive chemotherapy or healthy controls. The purpose of the current investigation was to assess gray matter alterations related to breast cancer and its treatment pro­ spectively in an independent cohort of patients treated with and without standard-dose systemic chemotherapy and demographi­ cally matched healthy controls, in order to replicate our previous findings. Given the prominence of executive function changes among the cognitive domains affected in cancer patients after treatment (Anderson-Hanley et al., 2003), and the recent finding of a relationship between self-reported executive functioning and altered brain activation after breast cancer chemotherapy (Kesler et al., 2011), we also sought to examine the relationship of these gray matter changes to self-reported executive functioning. Finally, a large body of research has shown a significant relationship be­ tween the apolipoprotein E (APOE) e4 allele and Alzheimer’s dis­ ease and its precursors, and has demonstrated a role for APOE in other neurocognitive disorders (for reviews see (Bookheimer and Burggren, 2009; Smith, 2000)). Given prior work demonstrating decreased cognitive functioning in cancer survivors treated with chemotherapy who carried the e4 allele vs. those who did not (Ahles et al., 2003), we further evaluated possible risk factors for gray matter changes after chemotherapy by investigating their relationship to presence or absence of the APOE e4 allele.