Associations of cognition, mood symptoms, and brain regional homogeneity in patients with breast cancer with or without chemotherapy and healthy controls

doi:10.21203/rs.3.rs-2335565/v1

Download PDF

Research Article

Associations of cognition, mood symptoms, and brain regional homogeneity in patients with breast cancer with or without chemotherapy and healthy controls

https://doi.org/10.21203/rs.3.rs-2335565/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

Background

Cognitive impairments have long been a complaint from patients with BC. We hypothesized that presentations on Regional homogeneity (ReHo) from fMRI may be associated with relevant cognitive changes in patients with or without breast cancer and/or chemotherapy.

Methods

Neuropsychological assessments of cognitive functions, levels of depression, fatigue, and anxiety, as well as whole-brain MRI scans were administered in patients with newly diagnosed BC prior to and 3~9 months after receiving chemotherapy, as well as healthy controls without cancer. ReHo was calculated from fMRI data to determine synchronizations of local brain activity. Multivariate regression models adjusting for intelligence quotient (IQ), menopause, and mood symptoms, as well as mediation analyses using generalized structural equation modeling, were performed.

Results

In all, 51 participants (19 noncancer controls, 11 patients with BC before chemotherapy, and 21 patients with BC who finished chemotherapy) completed the neuropsychological assessments and MRI. Significant differences in IQ and ReHo from several brain areas were observed in the three subgroups. Predictors for each domain of neurocognitive testing differed among the subgroups. Brain synchronization from the right middle frontal area was found to have significant mediating effect between chemotherapy status and the first part of Color Trails Test (CTT1) in the pre-C/T subgroup.

Conclusion

Differing from our expectations, the effects that the status of chemotherapy had on neurocognitive function assessed by CTT1 was mediated by ReHo in the right middle frontal area individually and not in sequential order with any mood symptoms.

Breast cancer

chemotherapy

cognition

MRI

Questions

Regional homogeneity in differed among BC patients with or without chemotherapy.
Whether brain synchronization mediated between cancer and cognitive functions.

Finding

The effects that BC had on neurocognitive function as measured using digit span were mediated by depression or ReHo in the right gyrus rectus area individually and not in sequential order.
The effect of cancer on the CTT1 was mediated by ReHo in the Right middle frontal lobe.

Meaning

The study both provides neurological evidence regarding cancer, chemotherapy, and cognitive functions and demonstrates how depression and brain synchronization play a mediating role between cancer and attention but not in other cognitive tests.

Although the worldwide incidence of breast cancer (BC) was as high as 2.3 million in 2020, by virtue of early detection programs and improvements in various treatment methods help decrease mortality rates in patients with breast cancer.(1) As the survival rate rises, issues regarding quality of life have become the focus of attention. Cognitive impairments have long been a complaint of patients with BC. Between 15% and 60% of patients complain cognitive decline that interferes with their quality of life.(2–6) Underlying reasons for cognitive complaints may be associated with baseline intelligence,(7, 8) psychosocial stressors, mood symptoms of depression, anxiety, fatigue,(5, 9, 10) or inflammatory responses.(11) Although adverse neurotoxic effects from systemic chemotherapy have also been suggested,(12) studies have reported that neurocognitive complaints are noticed before the commencement of chemotherapy in 21%~33% of patients with BC(10), and that cognitive decline might be associated not only with chemotherapy but also with the status of cancer.(9, 11)

Advancements in magnetic resonance imaging (MRI) technology have enabled the further exploration of factors associated with cognitive functions in BC patients.(10) Reductions in regional gray matter density(13), volume(14), and white matter integrity(15) associated with attention and memory impairments after chemotherapy have been shown in structural MRI research.(16, 17) Functional MRI studies have revealed decreased prefrontal activation(18) during working memory tasks 1 month after chemotherapy, but it returned to pretreatment hyperactivation (compared with healthy controls) 1 year after chemotherapy.(19) A study also described correlations among attention, executive functions, and strength of functional connectivity between dorsal lateral prefrontal cortex and contralateral cerebellar lobes.(20) The integration of imaging techniques investigating mechanisms of cognitive impairments in patients with BC remains an area requiring exploration.

Emerging evidence has indicated that cognitive complaints in patients with BC may not be as simple as the “chemo brain” theory. In particular, these cognitive impairments may also be related to cancer itself(11). Some findings have suggested that cognitive impairments in patients who received chemotherapy are less severe than those of patients who did not.(10, 19, 21) Interactions between different factors may also be involved. Relevant mechanisms or mediating paths among these potential influencing factors (between the status of cancer or chemotherapy and cognitive impairments) are lacking. Whether cognitive impairments in BC patients are affected by factors such as emotions, baseline demographic characteristics, treatments, or the interactions between factors may also be clarified through brain imaging techniques.

This study aimed to be exploratory to clarify possible associations of mood, and status of cancer and chemotherapy, as well as their effects to cognitive functions in BC patients using the MRI. We hypothesized that Regional homogeneity (ReHo) in different brain regions may be associated with relevant cognitive changes (22). ReHo is a measure of the temporal consistency signaled by similar blood oxygen levels in the nearby local brain tissues. It can be used to analyze abnormal brain neural activity. Excessive ReHo may represent overactivity or internal consonance of neurons in localized brain areas(23). Our conceptual model postulated that cancer or chemotherapy first causes emotional symptoms, such as depression, anxiety, or fatigue, and then affects cognitive functions through brain synchronization. The other direction—that cancer and chemotherapy first influence brain synchronizations, which then affect mood symptoms that lead to cognitive dysfunction—was also examined. Despite the assumption of sequential effects in mood and brain synchronization between cancer or chemotherapy and cognitive functions, these factors may also operate independently. Mediating factors and associated paths with significant direct or indirect effects were therefore explored.

Participants

We invited adult patients with Stage I to Stage III invasive BC without metastasis from the oncology department of a medical center in Southern Taiwan to participate. These patients included BC patients with (within 3 ~ 9 months of completing chemotherapy [post-C/T]) and without chemotherapy (pre-C/T group). Healthy volunteers without cancer within 5 years of the BC patients’ ages were also recruited. Our exclusion criteria were current pregnancy, illiterate, or history of severe visual or neurological impairment including epilepsy, stroke, multiple sclerosis, head injury with neurological sequela, movement disorder, or central nervous system lesions. This study was approved by the Research Ethics Committee of Chiayi Chang Gung Memorial Hospital (IRB number: 201700255B0).

Study procedure

After the patients provided consent, demographic variables, including job, education years, menopause, body mass index, comorbidities, and treatment receipts, were recorded. Levels of depression, anxiety, and fatigue were evaluated using the Taiwanese versions of the Patient Health Questionnaire (PHQ-9),(24) the anxiety subscale of the Hospital Anxiety and Depression Scale (HADS-A),(25) and the Brief Fatigue Inventory (BFI).(26) The reason for obtaining these variables was that they might have effects on cognitions and/or mood states. (9)

Neuropsychological measures

Neurological assessments were carried out by trained psychologists. The executive functions were assessed by the Semantic Association of Verbal Fluency Test [SFT],(27) Orthographical Fluency Test [OFT],(27) and Taiwanese Color Trails Test, second part [CTT2] (28). The first part of the Color Trails Test (CTT1) (28) was used to evaluate attention. The word list subtest of the Wechsler Memory Scale–Third Edition (Immediate or delayed recall, and recognition), as well as the Event- or Time- based Prospective Memory tests were used to assess memory(29, 30). IQ was used as a covariate and was evaluated by the Wechsler Adult Intelligence Scale- Third Edition(31). Better cognitive performance was indicated by higher scores on these measures, except on the CTT1 and CTT2, whose scores have an inverse relationship with cognitive function. Details regarding the psychometric validity of these assessments are described in our previous publication.(32)

Functional MRI acquisition

All participants were scanned using a 3T MRI (Verio, SIEMENS, Erlangen, Germany) system with a standard 8-channel head coil at Chiayi Chang Gung Memorial Hospital. A gradient- echo planar image sequence obtained resting-state functional images, with repetition time/echo time = 2000/30 ms, flip angle = 90°, number of excitations = 1, field of view = 220 × 220 mm², matrix size = 64 × 64, and voxel size = 3.4 × 3.4 × 4 mm³; 31 axial images were acquired to cover the entire brain volume. Each resting-state functional MRI run contained 300 image volumes, and the total scan time was approximately 10 min.

Functional image preprocessing

Functional MRI data preprocessing was performed using statistical parametric mapping 12 (SPM12; Wellcome Department of Cognitive Neurology, London, UK). Data were processed for slice-timing correction and then realigned to their first volume for motion correction. The results of six head motion parameters surpassed 1-mm translation or 1° rotation, and were excluded from this study. All participants fit the criteria without significant movements. After motion correction, data were normalized to standard Montreal Neurological Institute space with affine transformation, and the data were resampled to isotropic 3-mm voxels. Then, using the six head motion parameters as covariates, we performed nuisance regression. Then, the whole-brain, white matter, and cerebrospinal fluid (CSF) masks were used to remove the physiological noise. Linear detrending and band-pass temporal filtering were performed on the time series of each voxel to minimize the effects of low-frequency drifts and physiological signals by using the Resting-State Data Analysis tool kit v1.8 (REST v1.8, Center for Cognition and Brain Disorders, Hangzhou Normal University, Zhejiang, China). We used the frequency range from 0.01 to 0.12 Hz to observe the complex functional networks and mitigate the influence of low-frequency drift and high-frequency physiological noise.(33)

Reginal homogeneity

Regional homogeneity (ReHo) is based on the concept that BOLD signal fluctuations in a specific region reflect near neuronal activity arising at the same frequency and that this temporal synchrony is confined to groups of neurons that perform related functions.(22) ReHo analysis was conducted to determine the peak coordinates of brain regions with abnormal local brain activity synchronization. The identified brain regions were then considered seeds, and functional connectivity between these brain regions and global voxels was computed. After computation, the potential associations between these imaging indices and data from neuropsychological assessments were explored.(34)

Statistical analyses

To compare differences in demographic information, mood symptoms, and ReHo among BC patients with or without chemotherapy and healthy controls, a chi-squared test and analysis of variance (ANOVA) were applied for categorical and continuous variables or ReHo, as appropriate. Two-sample t-tests were performed for post hoc analysis.

Results from neuropsychological evaluations, levels of mood symptoms (anxiety, depression, or fatigue), menopause status, and brain synchronization (ReHo) that differ significantly in ANOVA were included in stepwise multivariable linear regression analyses to select potential covariates associated with each cognitive test for further mediation analysis. A p- value less than .05 was considered statistically significant.

To examine whether a multistage model explaining cognitive differences from significant variables of cancer or chemotherapy status, mood symptoms, or ReHo found in multivariable linear regression existed, mediation analyses using generalized structural equation modeling (GSEM) were performed.(35) Our structural model for GSEM hypothesized that the effect of cancer or chemotherapy status on neuropsychological functions was mediated or at least partially mediated by mood symptoms and/or brain synchronization (measured by ReHo) in certain brain areas. Menopause status was regarded as other correlated factor and was not on the mediation path. Confirmatory analyses on one or two (if mood symptoms and ReHo were significant variables) possible mediation pathways were examined using analysis of moment structure (AMOS version 16). The included pathways were as follows: Model 1: status of cancer or chemotherapy as the independent variable, each neurocognitive test as the dependent variable, mood symptoms as the first mediating variable, and ReHo as the second mediating variable; Model 2: same independent and dependent variables, with ReHo as the first mediating variable and mood symptoms as the second mediating variable. Path analyses were examined in the three subgroups together and separately to determine whether any path influenced the three subgroups differently. Estimated direct, indirect, and total effects between our predictors and outcomes were also calculated.

In all, 51 participants (19 noncancer controls, 11 BC patients that did not receive chemotherapy yet, and 21 patients that finished chemotherapy within 3~9 months) completed the neuropsychological assessments and MRI. Clinical and demographic characteristics and differences among the three subgroups are shown in Table 1. The mean age of the combined sample was 48.6 (SD = 11.2) years. The three groups were similar in demographic and mood status. Only the mean IQ in noncancer controls was significantly higher than that among patients in the prechemotherapy group (p = .002) but the education years did not differ significantly. Average scores on questionnaires of anxiety and depression were similar among the three groups, and all under the cutoff points for clinical depressive or anxiety disorders. Most of the patients with BC were diagnosed as having Stage II BC, and all patients in the post-C/T group underwent mastectomy. The main regimen of chemotherapy (in approximately 40% of the post-C/T subgroup) was 5-fluorouracil, cyclophosphamide, docetaxel, epirubicin, and doxorubicin. No significant differences in mean scores of the neuropsychological assessments were observed among the three subgroups (Table 1).

Significant differences in ReHo were observed in the left superior occipital area and right thalamus (the pre-C/T subgroup had lower ReHo than did the post-C/T subgroup), left angular (higher ReHo in the pre-C/T than in the noncancer subgroup), right middle frontal area and right frontal superior medial (the pre-C/T subgroup had higher ReHo than did the post-C/T subgroup), left frontal superior lobe (the pre-C/T subgroup had higher ReHo than did the other subgroups), left fusiform, and left and right calcarine (the pre-C/T subgroup had lower ReHo than did the other subgroups), left pallidum (the post-C/T subgroup had lower ReHo than did the noncancer subgroup), right rectus (the post-C/T subgroup had higher ReHo than did the noncancer subgroup), and left insula and left and right lingual (the pre-C/T subgroup had lower ReHo than the noncancer subgroup) (Table 1).

Stepwise multivariable regression analysis (in Table 2) revealed associations among mood symptoms, IQ, menopause status, ReHo, and each neuropsychological assessment. For attention function evaluated by CTT1 scores, ReHo from the superior left occipital lobe had significant positive associations in the noncancer and post-C/T subgroups. In the post-C/T subgroup, ReHo in the right middle frontal area was positively associated with CTT1, and IQ was negatively associated with CTT1 scores. In the pre-C/T subgroup, the left angular and left insula were positively and negatively associated with CTT1 score, respectively. In the domain of executive function, the SFT was negatively associated with ReHo in the right pallidum and right middle frontal area connectivity in the noncancer subgroup. In the post-C/T subgroup, the SFT was positively associated with IQ and anxiety, but negatively associated with ReHo in the right rectus area. The OFT was negatively associated with the right pallidum and left fusiform in the noncancer subgroup. In the pre-C/T subgroup, the OFT was negatively associated with the left fusiform, left angular, and right thalamus and with menopause and was positively associated with the left calcarine. Executive function, assessed by the CTT2, revealed negative associations with the left calcarine in the pre-C/T and with IQ in the post-C/T subgroups, respectively. With regard to memory function, word list immediate recall (WMS_Im) was positively associated with menopause status in the noncancer subgroup. WMS_Im was negatively associated with ReHo from the left superior occipital area and right lingual regions in the pre-C/T subgroup. A positive association between WMS_Im and ReHo from the left pallidum was observed in the post-C/T subgroup. For word list long delay recall (WMS_De), a positive association was observed with menopause in the noncancer subgroup. Negative associations between fatigue and the left superior occipital area were noted in the pre-C/T subgroup. For word list recognition (WMS_Re), a positive association with menopause status, and a negative association with the right thalamus were present in the noncancer subgroup. Negative associations of WMS_Re with fatigue and the right calcarine were observed in the pre-C/T subgroup. Positive associations of WMS_Re with the right calcarine and left superior occipital lobe were present in the post-C/T subgroup. A positive association was present between Event-based memory and the left calcarine in the noncancer subgroup. Negative associations for time-based memory, menopause, and IQ were observed in the noncancer subgroup (all in Table 2).

Table 3 contains the results of GSEM analyses comparing direct, indirect, and total effects from status of cancer or chemotherapy, mood symptoms, ReHo, and neurocognitive outcomes from Table 2 among the three subgroups. Among all the cognitive tests, cancer status was significantly associated with the memory function of WMS_Im (estimate = −1.679, p = .029) (CMIN/DF = 5.16, GFI = .84, NFI = .39, CFI = .35, RMSEA = .144), WMS_De (estimate = −1.62, p = .028) (CMIN/DF = 1.31, GFI = .99, NFI = .95, CFI = .98, RMSEA = .04), and recognition (estimate = −1.375, p = .047) (CMIN/DF = 2.61, GFI = .85, NFI = 0, CFI = 0, RMSEA = .09). However, none of these associations were mediated by mood or brain homogeneity. Menopause status had direct effects on the memory functions in WMS_Im (estimate = 2.161, p = .003), WMS_De (estimate = 2.416, p = .002), and recognition (estimate = 2.102, p = .004). A significant positive correlation was noted between menopause and cancer status (r = .099, SE = .035, p = .004). IQ was also observed to have significant direct effects on CTT1 (estimate = −0.916, p < .001), SFT (estimate = 0.346, p < .001), and CTT2 (estimate = −1.01, p = .004).

Table 4 presents the subanalysis of GSEM results comparing direct, indirect, and total effects for chemotherapy status, mood symptoms, ReHo, and neurocognitive outcomes between patients with and without chemotherapy. Chemotherapy status was significantly associated with CTT1 scores. A significant direct effect was present between chemotherapy status and CTT1 score (Table 4, estimate = 39.11, p < .001), and significant indirect effects were present through ReHo from the right middle frontal, left angular, and left superior occipital areas to the CTT1. However, only the mediation effects of the right middle frontal area between chemotherapy status and the CTT1 were significant. The model fit was within an acceptable range (CMIN/DF = 1.526, GFI = .867, NFI = .748, CFI = .873, RMSEA = .075). Figure 1 showed the aggregation of fMRI (ReHo) form the total sample in the right middle frontal area that showed significant differences from ANOVA analysis among the three subgroups. Paths for mediations between chemotherapy and CTT1 score are depicted in Figure 2. No other cognitive functions were associated with chemotherapy status. Menopause was significantly correlated with chemotherapy status (r= .103, SE=0.306, p=0.004). In this combined group of BC patients, IQ had significant direct effects on CTT1 (estimate = −1.72, P < .001), SFT (estimate = 0.486, p < .001), and CTT2 (estimate = −1.26, p = .004).

This was the first study to explore whether mood symptoms or brain synchronization exhibited in ReHo were possible mediators in associations between status of cancer or chemotherapy and cognitive functions. We observed that ReHo in several regions of the brain differed among BC patients with or without chemotherapy and a noncancer comparison group. In multivariable regression analyses, predictors for each domain of neurocognitive testing differed among the three subgroups. In mediation analyses, status of cancer or chemotherapy was significantly associated with WMS_Im, WMS_De, WMS_Re, or the CTT1, respectively. In the path analyses, brain synchronization from the right middle frontal area was found to have significant mediating effect between chemotherapy status and CTT1 in the pre-C/T subgroup.

Unlike our presumed conceptual model, mood symptoms and brain homogeneity did not have a sequential relationship between status of cancer or chemotherapy and cognitive functions. We presumed that cancer or chemotherapy might lead to mood changes, then alter brain homogeneity and cause cognitive functions among the three groups differ. We also speculated in the other direction that cancer or chemotherapy first altered brain homogeneity, then lead to mood changes and finally affected cognitive functioning. However, our results indicated individual rather than serial mediations in chemotherapy status and cognitive functions. The only significant mediator between chemotherapy and the CTT1 was ReHo from the right middle frontal area. This result demonstrated that in the pre-C/T subgroup, the longer time to complete the CTT1 (indicating poorer attention) was partially mediated by ReHo from the right middle frontal area. Past research on BC patients who received chemotherapy revealed associations between impaired executive functions and lower functional connectivity with the frontal cortex.(36) Increased difficulty in planning tasks was observed to be associated with prefrontal hyperactivation.(37) White matter integrity in the frontal, parietal, and occipital areas has also been related to changes in attention among patients that completed chemotherapy.(13, 16) Destructions in the basal forebrain and frontal lobes were theorized to impair the hippocampal system related to memory, or to interfere with neurotransmitter systems within the basal forebrain, such as cholinergic neurons or nearby catecholamine pathways.(38) The impacts and mechanisms on the attention ability in patients with cancer or chemotherapy may therefore not be single but have multiple aspects. Differences in brain homogeneity of the right middle frontal area might provide basis to clarify possible pathophysiological changes between chemotherapy and impaired attention in BC patients (36). Our results of negative indirect effects between attention and brain synchronization from the right middle frontal area may indicate the temporal consistency of hemodynamic responses in that brain area under the specific condition of chemotherapy. Such biological effects underlying brain homogeneity and cognitive deficits in BC patients still need further exploration.

From univariate analysis in Table 1, significant differences in brain ReHo were present in many but not all brain regions among the three subgroups. Although there might be too many variables in the stepwise multivariate regression analyses in the case of our modest sample size, the regression model was not the main analysis but played the role to help select associated specific brain regions, demographic variables, or mood states for further GSEM analysis. From the GSEM analysis, cancer status had significant direct negative effects on Wechsler Memory scales of Immediate, Delayed recall, and recognition (Table 3). This finding implied that having cancer might have possible impairments on memory function. However, the unsatisfactory model fits for WMS_Im and WMS_Re might indicate that our covariates were unable to fully explain associations between cancer and these memory tests. Our finding that scores on Wechsler memory word list subset and none of the other cognitive functions were significantly associated with chemotherapy status in Table 4 might support our findings from previous paper that cognitive dysfunctions may be associated with the status of cancer rather than chemotherapy.(11) In other cognitive tests, cancer status did not significantly affect the performances. Such result may indicate that although cancer status appeared to be significantly associated with ReHo in certain brain regions, it was not associated with and did not affect cognitive functions. Similarly, although mood conditions and ReHo in each brain region might have their effects on cognitive functions, since there were no significant effects between cancer and these cognitive tests, mood and ReHo were therefore not mediators between cancer status and cognitive tests. For instance, despite that anxiety and ReHo from the right gyrus rectus had significant direct effects on SFT, these effects did not make BC patients significantly susceptible to SFT impairments (as indicated by the nonsignificant association between cancer and the SFT in Table 3).

We demonstrated that IQ and menopause had significant effects on several cognitive tests, including IQ on the CTT1, the SFT, and the CTT2; and menopause on all three word list memory tests. The fact that higher IQ was associated with better cognitive performance is not surprising.(7, 8) Such findings are reminders of the need to include IQ and menopause status as baseline covariates in studies investigating cancer or chemotherapy and cognitive functions.(9, 11, 39) With regard to results associated with menopause status in mediation analyses of cancer status and cognitive tests, menopause had direct positive effects on WMS_Im, WMS_De, and WMS_Re; a negative direct effect on prospective time-based memory was also observed. In the circumstance of significant positive correlations between cancer or chemotherapy status and menopause, and having a large proportion of post-C/T patients with the status of menopause, we believe that menopause status might be correlated with cancer or its treatment. The menopause status among the three groups was originally quite different, mainly because the side effects of BC treatments may cause menopause. We examined menopause status separately was because collinearity may have occurred if menopause status was placed as a confounding factor. The negative direct effect that having menopause was associated with poorer performance on time-based prospective memory tasks was not surprising. The positive direct effect of menopause on better word list memory was difficult to explain. Although women often have subjective cognitive complaints after menopause, long-term research following up the effects of hormones on cognitive changes after menopause is still scant.(40) Most studies have focused on women over 65 years old,(41, 42) but the average age of our participants was 48.6 years. Furthermore, research has described that the domain for cognitive decline was a slight decrement in “learning ability,”(40–42) and not overall cognitive performance. Therefore, making inferences regarding these results for short- or long-term effects of menopause on cognition is difficult, particularly as the cognitive alterations in our patients may be cancer- related. Finally, hormonal changes may have different impacts on different brain functions. More research is required to explore these possible explanatory factors.

Strengths And Limitations

The main strength of the study was applying well-validated neuropsychological evaluation tools administered by trained psychologists to compare differences in cognitive functions, mood symptoms, and patient characteristics. Using whole-brain MRI to explore the relationships between brain homogeneity and cognitive functions in BC patients with or without chemotherapy and in noncancer comparisons was unprecedented. Major limitations are, first, our sample size may be too small, particularly the size of the pre-C/T subgroup. The small sample size affected our capability to obtain a favorable model fit (which is highly sensitive to sample size) from our mediation pathways. However, this is an exploratory study that aimed to investigate whether cancer or chemotherapy may influence cognitive functions mediated by changes in blood oxygen levels that affect brain homogeneity. To the least extent, the model fits for the significant path we found were within acceptable ranges, although increasing the sample size may help provide higher validity. Second, our cross-sectional design made comparing changes in mood states or cognitive functions before and after chemotherapy in the same participant unlikely. Long-term follow-up studies are required to determine whether cognitive deficits are reversible. Third, although we obtained extensive objective and subjective assessments to evaluate cognitive functions and states of mood and fatigue, objective psychiatric diagnostic interviews confirming whether participants actually had clinical syndromes of depression or anxiety were not conducted. Data for potential confounders associated with cancer and cognition, such as diet and exercise habits, were also unavailable. More detailed confirmations to avoid possible confounders, such as migraine as a side effect of hormonal therapy, molecular subtypes of breast cancer, or pre-surgery tumor volume, may still be needed. Finally, although the measurement tools used are internationally available, our results may only represent the cognitive functions of the Mandarin-speaking population in Taiwan. Restrictions on generalizability to other countries or ethnicities must be considered.

Unlike our expectations, the effects that BC had on memory were not mediated by mood or brain synchronizations. In the pre-C/T subgroup, the effect of cancer on the CTT1 was mediated by ReHo in the right middle frontal area. The study provides neurological evidence regarding cancer, chemotherapy, and cognitive functions and demonstrates how brain synchronization in the right middle frontal area may play a mediating role between chemotherapy and attention. Studies on the possible mechanisms of how intrinsic brain hemodynamics in the right middle frontal areas might influence CTT1 scores are warranted. Regarding relevance for health care professionals, our result may help to uncover interactions and mediating paths among potential influencing factors of baseline intelligence, mood symptoms, treatments, and cognitive functions. Maybe such knowledge may help reduce BC patients’ fear or worry of the less- studied cognitive impairments associated with cancer treatment.

Ethics approval and consent to participate

This study was approved by the Research Ethics Committee of Chiayi Chang Gung Memorial Hospital (IRB number: 201700255B0).

Consent for publication

Not applicable.

Availability of data and materials

Datasets being analyzed and results being generated and reported in this article are not publically available due to protections of personal privacy. Restrictions applied to these data, which were used under license for our study, and so are not publicly available for duplication. Further data analysis may be requested after discussion with authors.

Competing interests

The authors declare that they have no competing interests.

Funding

SIW is part-funded by Department of Medical Research, Mackay Memorial Hospital (MMH-109112, MMH-10914, MMH-108121, MMH-108146, MMH-TT-10804, MMH-TH-10804). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. This manuscript was edited by Wallace Academic Editing.

Authors' contributions

All authors have contributed significantly, and all authors are in agreement with the content of the manuscript: Conception/design: Vincent Chin-Hung Chen, Jun-Cheng Weng, and Shu-I Wu; Collection and/or assembly of data: All authors; Data analysis and interpretation: All authors; Manuscript writing: Shu-I Wu, Vincent Chin-Hung Chen, and Yen-Hsuan Hsu; Final approval of manuscript: All authors.

Acknowledgements

Not applicable.

WHO. Breast cancer 2021 [Available from: https://www.who.int/news-room/fact-sheets/detail/breast-cancer.
Reid-Arndt SA, Hsieh C, Perry MC. Neuropsychological functioning and quality of life during the first year after completing chemotherapy for breast cancer. Psycho-oncology. 2010;19(5):535-44.
Wu LM, Amidi A. Cognitive impairment following hormone therapy: current opinion of research in breast and prostate cancer patients. Current opinion in supportive and palliative care. 2017;11(1):38-45.
Yao C, Bernstein LJ, Rich JB. Executive functioning impairment in women treated with chemotherapy for breast cancer: a systematic review. Breast cancer research and treatment. 2017;166(1):15-28.
Ahles TA. Brain vulnerability to chemotherapy toxicities. Psycho-oncology. 2012;21(11):1141-8.
Wefel JS, Saleeba AK, Buzdar AU, Meyers CA. Acute and late onset cognitive dysfunction associated with chemotherapy in women with breast cancer. Cancer. 2010;116(14):3348-56.
Vearncombe KJ, Rolfe M, Wright M, Pachana NA, Andrew B, Beadle G. Predictors of cognitive decline after chemotherapy in breast cancer patients. Journal of the International Neuropsychological Society : JINS. 2009;15(6):951-62.
Wefel JS, Lenzi R, Theriault RL, Davis RN, Meyers CA. The cognitive sequelae of standard-dose adjuvant chemotherapy in women with breast carcinoma: results of a prospective, randomized, longitudinal trial. Cancer. 2004;100(11):2292-9.
Hsu YH, Chen VC, Hsieh CC, Weng YP, Hsu YT, Hsiao HP, et al. Subjective and objective cognitive functioning among patients with breast cancer: effects of chemotherapy and mood symptoms. Breast cancer (Tokyo, Japan). 2021;28(1):236-45.
Hermelink K. Chemotherapy and Cognitive Function in Breast Cancer Patients: The So-Called Chemo Brain. Journal of the National Cancer Institute Monographs. 2015;2015(51):67-9.
Chen VC, Lin CK, Hsiao HP, Tzang BS, Hsu YH, Wu SI, et al. Effects of Cancer, Chemotherapy, and Cytokines on Subjective and Objective Cognitive Functioning Among Patients with Breast Cancer. Cancers (Basel). 2021;13(11).
Schagen SB, Muller MJ, Boogerd W, Mellenbergh GJ, van Dam FS. Change in cognitive function after chemotherapy: a prospective longitudinal study in breast cancer patients. Journal of the National Cancer Institute. 2006;98(23):1742-5.
McDonald BC, Conroy SK, Smith DJ, West JD, Saykin AJ. Frontal gray matter reduction after breast cancer chemotherapy and association with executive symptoms: a replication and extension study. Brain, behavior, and immunity. 2013;30 Suppl:S117-25.
Koppelmans V, Breteler MM, Boogerd W, Seynaeve C, Gundy C, Schagen SB. Neuropsychological performance in survivors of breast cancer more than 20 years after adjuvant chemotherapy. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2012;30(10):1080-6.
de Ruiter MB, Reneman L, Boogerd W, Veltman DJ, Caan M, Douaud G, et al. Late effects of high-dose adjuvant chemotherapy on white and gray matter in breast cancer survivors: converging results from multimodal magnetic resonance imaging. Human brain mapping. 2012;33(12):2971-83.
Deprez S, Amant F, Smeets A, Peeters R, Leemans A, Van Hecke W, et al. Longitudinal assessment of chemotherapy-induced structural changes in cerebral white matter and its correlation with impaired cognitive functioning. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2012;30(3):274-81.
Ahles TA, Root JC, Ryan EL. Cancer- and cancer treatment-associated cognitive change: an update on the state of the science. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2012;30(30):3675-86.
Kesler SR, Kent JS, O'Hara R. Prefrontal cortex and executive function impairments in primary breast cancer. Archives of neurology. 2011;68(11):1447-53.
McDonald BC, Conroy SK, Ahles TA, West JD, Saykin AJ. Alterations in brain activation during working memory processing associated with breast cancer and treatment: a prospective functional magnetic resonance imaging study. Journal of clinical oncology : official journal of the American Society of Clinical Oncology. 2012;30(20):2500-8.
Park HY, Lee H, Sohn J, An SK, Namkoong K, Lee E. Increased resting-state cerebellar-cortical connectivity in breast cancer survivors with cognitive complaints after chemotherapy. Scientific reports. 2021;11(1):12105.
Pomykala KL, Ganz PA, Bower JE, Kwan L, Castellon SA, Mallam S, et al. The association between pro-inflammatory cytokines, regional cerebral metabolism, and cognitive complaints following adjuvant chemotherapy for breast cancer. Brain imaging and behavior. 2013;7(4):511-23.
Zang Y, Jiang T, Lu Y, He Y, Tian L. Regional homogeneity approach to fMRI data analysis. NeuroImage. 2004;22(1):394-400.
Xu Z, Lai J, Zhang H, Ng CH, Zhang P, Xu D, et al. Regional homogeneity and functional connectivity analysis of resting-state magnetic resonance in patients with bipolar II disorder. Medicine. 2019;98(47):e17962.
Liu SI, Yeh ZT, Huang HC, Sun FJ, Tjung JJ, Hwang LC, et al. Validation of Patient Health Questionnaire for depression screening among primary care patients in Taiwan. Compr Psychiatry. 2011;52(1):96-101.
Wang GL, Hsu SH, Feng AC, Chiu CY, Shen JF, Lin YJ, et al. The HADS and the DT for screening psychosocial distress of cancer patients in Taiwan. Psycho-oncology. 2011;20(6):639-46.
Lin CC, Chang AP, Chen ML, Cleeland CS, Mendoza TR, Wang XS. Validation of the Taiwanese version of the Brief Fatigue Inventory. Journal of pain and symptom management. 2006;32(1):52-9.
Hua MS, Chang SH, Chen ST. Factor structure and age effects with an aphasia test battery in normal Taiwanese adults. Neuropsychology. 1997;11(1):156-62.
Yang KH. HMS. Color Trails Test (Chinese). Taipei.: Chinese Behavioral Science Corp.; 2015.
Hua M, Chang, B.S., Lin, K.N., Yang, J.M, Lu, S.R., Chen S.Y. . Wechsler Memory Scale - Third Edition (Chinese), . Taipei.: Chinese Behavioral Science Corp.; 2005.
Hsu Y, Huang, CF., Tu, MC, Hua, MS. Prospective memory in subjective cognitive decline: A preliminary study on the role of early cognitive marker in dementia,. Alzheimer Dis Assoc Disord 2015;29:229-35.
Chen J, Chen, HY. . Wechsler Adult Intelligence Scale - Third Edition (Chinese). . Taipei, Taiwan.: Chinese Behavioral Science Corp., Taipei, 2002.; 2002.
Hsu YH, Chen, V.C., Hsieh, C.C., Weng, Y.P., Hsu, Y.T., Hsiao, H.P., Wang, W.K., Chen, H.M., Weng, J.C., Wu, S.I. . Subjective and objective cognitive functionings among patients with breast cancer: Eeffects of chemotherapy and mood symptoms. Supportive Care in Cancer. 2020.
Baria AT, Baliki MN, Parrish T, Apkarian AV. Anatomical and functional assemblies of brain BOLD oscillations. The Journal of neuroscience : the official journal of the Society for Neuroscience. 2011;31(21):7910-9.
Liu D, Duan S, Zhang J, Zhou C, Liang M, Yin X, et al. Aberrant Brain Regional Homogeneity and Functional Connectivity in Middle-Aged T2DM Patients: A Resting-State Functional MRI Study. Frontiers in human neuroscience. 2016;10:490.
Cole DA, Maxwell SE. Testing mediational models with longitudinal data: questions and tips in the use of structural equation modeling. J Abnorm Psychol. 2003;112(4):558-77.
Miao H, Li J, Hu S, He X, Partridge SC, Ren J, et al. Long-term cognitive impairment of breast cancer patients after chemotherapy: A functional MRI study. European journal of radiology. 2016;85(6):1053-7.
Menning S, de Ruiter MB, Veltman DJ, Koppelmans V, Kirschbaum C, Boogerd W, et al. Multimodal MRI and cognitive function in patients with breast cancer prior to adjuvant treatment--the role of fatigue. NeuroImage Clinical. 2015;7:547-54.
Damasio AR, Graff-Radford NR, Eslinger PJ, Damasio H, Kassell N. Amnesia following basal forebrain lesions. Archives of neurology. 1985;42(3):263-71.
Ahles TA, Saykin AJ. Breast cancer chemotherapy-related cognitive dysfunction. Clinical breast cancer. 2002;3 Suppl 3:S84-90.
Morgan KN, Derby CA, Gleason CE. Cognitive Changes with Reproductive Aging, Perimenopause, and Menopause. Obstetrics and gynecology clinics of North America. 2018;45(4):751-63.
Fuh JL, Wang SJ, Lee SJ, Lu SR, Juang KD. A longitudinal study of cognition change during early menopausal transition in a rural community. Maturitas. 2006;53(4):447-53.
Greendale GA, Huang MH, Wight RG, Seeman T, Luetters C, Avis NE, et al. Effects of the menopause transition and hormone use on cognitive performance in midlife women. Neurology. 2009;72(21):1850-7.

Table 1. Summary of participant characteristics
	Breast	Breast	Non-cancer	ANOVA
	cancer patients before chemotherapy (Pre C/T)	cancer patients after chemotherapy (Post C/T)	controls
	(N = 11)	(N = 21)	(N = 19)
Basic information	Mean (SD)	Mean (SD)	Mean (SD)	p	*Post-hoc test*
Age (years)	49.82 (14.20)	49.62 (11.11)	46.84 (10.46)	0.699
BMI	24.10 (4.62)	25.21 (4.84)	24.13 (2.69)	0.649
Education (years)	11.55 (3.75)	12.57 (3.28)	13.42 (3.50)	0.369
IQ^†	97.10 (7.37)	104.98 (12.77)	110.90 (7.42)	0.003*	control>pre-C/T
PHQ-9	4.18 (3.06)	3.33 (2.65)	1.89 (2.47)	0.068
HADS-A	3.36 (3.23)	2.57 (2.73)	2.21 (2.66)	0.56
BFI	0.81 (1.13)	1.21 (1.48)	1.32 (1.67)	0.668
	N (%)	N (%)	N (%)
Stage				<0.001
Stage I	2 (18.18)	7 (33.33)	-
Stage II	5 (45.45)	9 (42.86)	-
Stage III	4 (36.36)	4 (19.05)	-
Stage IV	0	1 (5.76)
Menopause (Yes)	6 (54.54)	21 (100.0)	8 (42.11)	0.001
Treatment
Mastectomy	4 (36.36)	21 (100.0)	-	<0.001
Radiotherapy	0 (0.0)	17 (81.0)	-	<0.001
Hormonal therapy	0 (0.0)	16 (76.19)	-	<0.001
Targeted therapy	0	0	-
Chemotherapy regimen
1.Cyclophosphamide, 5-fluorouracil, epirubicin and doxorubicin, and docetaxel	-	10 (47.62)	-
2.Cyclophosphamide, 5-fluorouracil, epirubicin and doxorubicin, docetaxel, and cisplatin	-	3 (14.29)	-
3. Cyclophosphamide, 5-fluorouracil, epirubicin and doxorubicin	-	2 (9.52)	-
4.Cyclophosphamide, methotrexate, and 5-fluorouracil	-	2 (9.5)	-
5.Cyclophosphamide, methotrexate, epirubicin and doxorubicin	-	4 (19.05)	-
Attention function
Color Trails Test 1	43.26 (14.99)	54.38 (30.97)	45.20 (14.86)	0.32
Executive function
Semantic Association of Verbal Fluency	40.36 (6.01)	38.71 (9.46)	44.32 (5.70)	0.069
Orthographical Fluency Test	19.82 (7.52)	20.81 (8.07)	21.91 (5.45)	0.729
Color Trails Test 2	92.69 (42.21)	93.61 (30.26)	81.91 (19.60)	0.428
Memory function
Word List - Total immediate recall	9.73 (2.33)	10.24 (2.84)	11.00 (2.65)	0.426
Word List - Long-delay recall	10.27 (2.10)	11.05 (2.44)	11.26 (2.94)	0.589
Word List - Recognition	11.18 (2.86)	10.96 (2.55)	11.47 (2.20)	0.948
Prospective memory
Event-based	3.91 (0.30)	4.00 (0.00)	3.79 (0.54)	0.184
Time-based	3.64 (0.50)	3.24 (0.77)	3.58 (0.77)	0.217
ReHo (only regions with significant differences among the three groups were presented here)
Left Angular	1.54 (0.19)	1.43 (0.14)	1.39 (0.13)	0.027	Pre C/T >Non-cancer
Left Calcarine	0.98 (0.12)	1.28 (0.27)	1.20 (0.20)	0.003	Pre- C/T < Post- C/T
					Pre- C/T < Non-cancer
Right Calcarine	1.00 (0.08)	1.20 (0.14)	1.18 (0.22)	0.008	Pre- C/T <Post- C/T
					Pre-C/T < Non- cancer
Right middle frontal lobe	1.15 (0.07)	1.04 (0.08)	1.10 (0.08)	0.003	Pre- C/T >Post- C/T
Left superior Frontal	1.15 (0.05)	1.03 (0.09)	1.06 (0.09)	0.001	Pre- C/T > Non-cancer
					Pre- C/T >Post- C/T
Right superior frontal medial	1.15 (0.08)	1.02 (0.14)	1.07 (0.13)	0.026	Pre- C/T >Post- C/T
Left Fusiform	0.83 (0.06)	0.92 (0.07)	0.91 (0.09)	0.005	Pre- C/T <Post- C/T
					Pre- C/T <Non-cancer
Left Insula	0.83 (0.06)	0.87 (0.07)	0.90 (0.05)	0.021	Pre- C/T <Non-cancer
Left Lingual	0.97 (0.17)	1.14 (0.16)	1.13 (0.18)	0.024	Pre- C/T <Non-cancer
Right Lingual	0.96 (0.16)	1.11 (0.16)	1.13 (0.19)	0.039	Pre- C/T <Non-cancer
Left superior occipital lobe	0.95 (0.07)	1.08 (0.14)	0.99 (0.12)	0.022	Pre- C/T <Post-C/T
Left Pallidum	0.74 (0.05)	0.72 (0.06)	0.79 (0.10)	0.011	Post- C/T <Non-cancer
Right Pallidum	0.84(0.10)	0.86 (0.09)	0.94 (0.13)	0.027
Right Putamen	0.83 (0.04)	0.85 (0.06)	0.89 (0.08)	0.042
Right Rectus	1.00 (0.17)	1.06 (0.19)	0.92 (0.14)	0.034	Post- C/T >Non-cancer
Right Thalamus	0.77 (0.06)	0.90 (0.13)	0.86 (0.10)	0.01	Pre- C/T <Post- C/T
† Intelligence quotient (IQ) estimated using the short form of the Taiwan Wechsler Adult Intelligence Scale-III; PHQ-9: Patient Health Questionnaire-9; HADS-A: Hospital Anxiety Depression Scale; BFI: Brief Fatigue Inventory; ‡ Significant difference after Bonferroni correction;

Table 2. Stepwise multivariate regression analysis of cognitive functions, menopause, mood symptoms, and ReHo

	Non cancer (n=21)					Pre-C/T (n=13)					Post-C/T (n=21)
Cognitions and Predictors	B	R²	ΔR²	p-value	Confidence interval (CI)	B	R²	ΔR²	p-value	CI	B	R²	ΔR²	p-value	CI
Color Trails Test 1		0.21	0.21	0.049				0.848	0.289	0.002				0.609	0.093	0.044
Left superior occipital lobe	55.232			0.049	0.395 ~	110.07							75.16			0.044	2.28 ~	148.05
Left Angular							52.98			0.001	29.43 ~	76.52
Left Insula							-133.04			0.002	-203.39 ~	-62.69
Right middle frontal lobe													188.24			0.007	59.51 ~	316.96
IQ													-1.95			<0.001	-2.7 ~	-1.21
Semantic Association of Verbal Fluency		0.44	0.225	0.022										0.733	0.161	0.003
Right pallidum	-28.56			0.006	-47.64 ~	-9.48
Right middle frontal lobe	-34.23			0.022	-62.86 ~	-5.60
Right gyrus rectus													-21.05			0.003	-33.85 ~	-8.25
Anxiety													1.51			0.002	0.65 ~	2.37
IQ													0.49			<0.001	0.3 ~	0.68
Orthographical Fluency Test		0.344	0.204	0.031				0.93	0.09	0.012
Right pallidum	-21.17			0.022	-38.82 ~	-3.52
Left fusiform	-28.17			0.031	-53.40 ~	-2.93	-1.74			0.005	-2.80 ~	-0.68
Left angular							-28.49			<0.001	-38.70 ~	-18.27
Left Calcarine							27.75			0.006	11.11 ~	44.19
Right thalamus							-68.78			0.003	-103.55 ~	-34.00
Menopause							-5.36			0.012	-9.03 ~	-1.70
Color Trails Test 2								0.583	0.583	0.006				0.436	0.436	0.001
Left Calcarine							-261.48			0.006	-428.10 ~	-94.86
IQ													-1.57			0.001	-2.42 ~	-0.71
Word List - Total immediate recall		0.386	0.386	0.005				0.903	0.052	0.05				0.191	0.191	0.047
Menopause	3.24			0.005	1.15 ~	5.33
Left superior occipital lobe							-26.62			<0.001	-34.86 ~	-18.38
Right Lingual							-3.68			0.05	-7.36 ~	-0.01
Left pallidum													21.21			0.047	0.26 ~	42.16
Word List - Long-delay recall		0.36	0.4	0.004				0.79	0.21	0.014
Menopause	3.65			0.004	1.34 ~	5.95
Left superior occipital lobe							-23.21			0.001	-33.07 ~	-13.36
Fatigue							-0.85			0.014	-1.48 ~	-0.22
Word List - Recognition		0.52	0.3	0.006				0.914	0.17	0.002				0.39	0.39	0.002
Menopause	2.51			0.006	0.83 ~	4.2
Right thalamus	-13.93			0.002	-22.10 ~	-5.77
Right calcarine							-32.77			<0.001	-40.24 ~	-17.34	22.53			0.044	0.67 ~	44.38
Left superior occipital lobe													10.56			0.002	4.22 ~	16.9
Fatigue							-1.06			0.002	-1.62 ~	-0.51
Event-based		0.258	0.258	0.026
Left calcarine	1.34			0.026	-2.51 ~	-0.18
Time-based		0.568	0.133	0.042
Menopause	-1.08			0.001	-1.61 ~	-0.54
IQ	-0.038			0.042	-0.075 ~	-0.002
Dependent variables: cognitive tests. Covariates: IQ, menopause, depression, anxiety, fatigue, and brain areas that had significant differences (left angular, left and right calcarine, right frontal middle, left frontal superior, left fusiform, left insula, left and right lingual, Left superior occipital lobe, left and right pallidum, right putamen, right rectus, and right thalamus.

Table 3. Direct, indirect, and total effects of cancer status, cognitive measures, IQ, menopause, mood symptoms, and ReHo

Model path	Effects
	Direct		Indirect	Total
Color Trails Test 1 (CTT1)
Cancer or not -> Left angular	0.079			0.079
Cancer or not ->Left Insula	-0.044	*		-0.044
Cancer or not -> Right frontal mid	-0.019			-0.019
Cancer or not -> Left occipital sup	0.042			0.042
Cancer or not -> CTT1	-6.887		4.335	-2.552
Left angular -> CTT1	1.235			1.235
Left Insula -> CTT1	-62.127			-62.127
Right frontal mid -> CTT1	19.34			19.34
Left occipital sup -> CTT1	44.81			44.812
IQ -> CTT1	-0.916	**		-0.916
Semantic association of verbal fluency (SFT)
Model 1
Cancer or not-> Right pallidum	-0.085	**		-0.085
Cancer or not -> Right frontal mid	-0.017			-0.019
Cancer -> Right rectus	0.131	**		0.131
Cancer or not-> Anxiety	0.63		0.001	0.633
Anxiety -> Right pallidum	0
Anxiety -> Right frontal mid	-0.003			-0.003
Anxiety -> Right rectus	-0.016	*		-0.016
Cancer or not-> SFT	-1.57		-0.486	-2.054
Right pallidum -> SFT	-10.637			-10.637
Right frontal mid -> SFT	-3.716			-3.716
Right rectus -> SFT	-15.594	*		-15.594
Anxiety-> SFT	0.662		0.271	0.934
IQ-> SFT	0.346	*******		0.346
Model 2
Cancer or not-> Right pallidum	-0.085	**		-0.085
Cancer or not -> Right frontal mid	-0.019			-0.019
Cancer -> Right rectus	0.121	**		0.121
Cancer or not-> Anxiety	0.994		-0.36	0.633
Right pallidum >anxiety	-1.915			-1.915
Right frontal mid-> anxiety	-2.118			-1.915
Right rectus -> anxiety	-4.674	*		-4.674
Cancer or not-> SFT	-1.57		-0.486	-2.054
Right pallidum -> SFT	-10.637		-1.268	-11.905
Right frontal mid -> SFT	-3.716		-1.402	-5.118
Right rectus -> SFT	-15.594	*	-3.094	-18.689
Anxiety-> SFT	0.662	*	0.271	0.662
IQ-> SFT	0.346	**		0.346
Orthographical Fluency Test (OFT)
Cancer or not -> Left calcarine	-0.025			-0.025
Cancer or not -> Left Angular	0.079			0.079
Cancer or not -> Right pallidum	-0.085	*		-0.085
Cancer or not -> OFT	0.717		-0.468	0.249
Left fusiform -> OFT	-0.531			-0.531
Right thalamus -> OFT	-17.991			-17.991
Right pallidum -> OFT	-8.639			-8.639
Left calcarine -> OFT	0.948			0.948
Left angular -> OFT	-14.555			-14.555
Menopause -> OFT	-3.999			-3.999
Color Trails Test 2 (CTT2)
Cancer or not -> Left calcarine	-0.025			-0.025
Cancer or not -> CTT2	2.369		0.305	2.674
Left calcarine -> CTT2	-12.136			-12.136
IQ-> CTT2	-1.01	**		-1.01
Wechsler Memory Scale- Immediate recall (WMS_Im)
Cancer or not -> Right lingual	-0.070			-0.070
Cancer or not -> Left pallidum	-0.069	**		-0.069
Cancer or not -> Left occipital sup	0.042			0.042
Cancer or not -> WMS_Im	-1.679	*	-0.172	-1.851
Right Lingual -> WMS_Im	-1.916			-1.916
Left pallidum -> WMS_Im	6.555			6.555
Left occipital sup -> WMS_Im	3.587			3.587
Menopause -> WMS_Im	2.161	**		2.161
WMS- Delayed recall (WMS_De)
Cancer or not -> Left Occipital sup	0.027			0.029
Fatigue -> Left occipital sup	-0.005			-0.005
Cancer -> Fatigue	-0.238			-0.238
Cancer or not -> WMS_De	-1.62	*	0.093	-1.526
Left occipital sup -> WMS_De	1.775			1.775
Fatigue -> WMS_De	-0.178		-0.01	-0.188
Menopause -> WMS_De	2.416	**	0.055	2.471
Cancer or not -> Left Occipital sup	0.027			0.027
Menopause -> left occipital	0.034			0.034
Left occipital sup-> fatigue	-0.668		0.119	-0.668
Menopause -> fatigue	-0.539		-0.023	-0.561
Cancer -> Fatigue	0.018		-0.018	-0.238
Cancer or not -> WMS_De	-1.62	*	0.049	-1.571
Left occipital sup -> WMS_De	1.775			1.894
Fatigue -> WMS_De	-0.178		-0.01	-0.178
Menopause -> WMS_De	2.416	**	0.16	2.576
WMS- Recognition test (WMS_Re)
Model 1
Cancer or not -> right calcarine	-0.048		-0.005	-0.053
Fatigue -> Right calcarine	0.020			0.02
Menopause -> right calcarine	0.001			0.001
Cancer or not -> Left occipital sup	0.027		0.001	0.029
Fatigue-> Left occipital sup	-0.005			-0.005
Menopause-> Left occipital sup	0.031			0.031
Cancer -> Right thalamus	-0.026			-0.026
Fatigue -> Right thalamus	0.000
Menopause -> Right thalamus	0.067			0.067
Cancer -> fatigue	-0.238			-0.238
Cancer or not -> WMS_Re	-1.375	*		-1.375
Right calcarine -> WMS_Re	-1.81			-1.81
Left occipital sup -> WMS_Re	6.878	*		6.878
Right thalamus -> WMS_Re	-6.222	*		-6.222
Fatigue -> WMS_Re	0.044		-0.073	-0.030
Menopause -> WMS_Re	2.102	*	-0.206	1.896
Model 2
Cancer or not -> right calcarine	-0.048			-0.048
Right calcarine -> fatigue	2.518	*		2.518
Menopause -> right calcarine	-0.01			-0.01
Cancer or not -> Left occipital sup	0.027			0.027
Left occipital sup-> Fatigue	-2.385			-2.385
Menopause-> Left occipital sup	0.034			0.034
Cancer -> Right thalamus	-0.026			-0.026
Right thalamus -> fatigue	-0.914			-0.914
Menopause -> Right thalamus	0.067			0.067
Cancer -> fatigue	-0.004		-0.162	-0.167
Cancer or not -> WMS_Re	-1.375	*	0.429	-0.946
Right calcarine -> WMS_Re	-1.81		0.111	-1.699
Left occipital sup -> WMS_Re	6.878	*	-0.105	6.773
Right thalamus -> WMS_Re	-6.222		-0.040	-6.222
Fatigue -> WMS_Re	0.044		-0.073	0.044
Menopause -> WMS_Re	2.102	*	-0.172	1.93
Prospective memory- Event based
Cancer or not -> Left calcarine	-0.025			-0.025
Cancer -> event based	0.173		0.006	0.179
Left calcarine -> Event based	-0.257			-0.257
Prospective memory- Time based
Cancer or not -> Time based	0.123			0.123
Menopause -> Time based	-0.765	*******		-0.765
IQ-> Time based	0.000			0.000
The model included cancer status, IQ, menopause, anxiety, depression, fatigue, and brain areas that were significantly different among the 3 groups in Tables 1 &2. Model 1: status of cancer or chemotherapy as the independent variable, each neurocognitive test as the dependent variable, mood symptoms as the first mediating variable, and ReHo as the second mediating variable; Model 2: same independent and dependent variables, with ReHo as the first mediating variable and mood symptoms as the second mediating variable. Bold letters: the effect was statistically significant: p<0.05, p<0.01, **p<0.001.

Table 4. Subanalysis of direct, indirect, and total effects of chemotherapy status

Model path	Effects
	Direct		Indirect	Total
Color Trails Test 1 (CTT1)
Chemo or not -> Left angular	-0.114	*		-0.114
Chemo or not ->Left Insula	0.032			0.032
Chemo or not -> Right middle frontal area	-0.108	*******		-0.108
Chemo or not -> Left occipital sup	0.122	**		0.122
Chemo or not -> CTT1	39.111	*******	-14.406	24.704
Left angular -> CTT1	42.13	*
Left Insula -> CTT1	-128.622**			-128.622**
Right middle frontal area -> CTT1	91.034	*		91.034
Left occipital sup -> CTT1	36.209			36.209
IQ -> CTT1	-1.72	*******		-1.72
Semantic Association of Verbal Fluency (SFT)
Model 1
Chemo or not-> Right pallidum	0.014			0.014
Chemo or not -> Right middle frontal area	-0.111	**		-0.108
Chemo -> Right rectus	0.049		0.003	0.06
Chemo or not-> Anxiety	-0.792		0.001	-0.792
Anxiety -> Right pallidum	-0.001			-0.001
Anxiety -> Right middle frontal area	-0.003			-0.004
Anxiety -> Right rectus	-0.014			-0.014
Chemo or not-> SFT	-4.766		-0.718	-5.484
Right pallidum -> SFT	8.462			8.462
Right middle frontal area -> SFT	-8.556			-8.556
Right rectus -> SFT	-17.316	*		-17.316
Anxiety-> SFT	0.914	*	0.27	1.185
IQ-> SFT	0.486	*******		0.486
Model 2
Chemo or not-> Right pallidum	0.014			0.014
Chemo or not -> Right frontal mid	-0.108	**		-0.108
Chemo -> Right rectus	0.06			0.06
Chemo or not-> Anxiety	-0.934		0.142	-0.792
Right pallidum >anxiety	-1.340			-1.34
Right middle frontal area -> anxiety	-3.344			-3.344
Right rectus -> anxiety	-4.674	*		-3.348
Chemo or not-> SFT	-4.766		-0.718	-5.484
Right pallidum -> SFT	8.462		-1.225	7.237
Right middle frontal area -> SFT	-8.556		-3.057	-11.613
Right rectus -> SFT	-17.316	**	-3.061	-20.377
Anxiety-> SFT	0.914	*	0.271	0.914
IQ-> SFT	0.486	**		0.486
Orthographical Fluency Test (OFT)
Chemo or not -> Left calcarine	0.302	**		4.816
Chemo or not -> Left Angular	-0.114	*		-17.289
Chemo or not -> Right thalamus	0.127	**		0.127
Chemo or not -> Left fusiform	0.09	**		0.09
Chemo or not -> Right pallidum	0.014			0.014
Chemo or not -> OFT	5.726		-1.274	4.452
Left fusiform -> OFT	0.564			0.564
Right thalamus -> OFT	-37.892	*		-37.892
Right pallidum -> OFT	2.649			2.649
Left calcarine -> OFT	4.816			0.948
Left angular -> OFT	-17.289			-14.555
Menopause -> OFT	-7.614			-7.614
Color Trails Test 2 (CTT2)
Chemo or not -> Left calcarine	0.302	**		0.302
Chemo or not -> CTT2	20.114		-9.249	10.865
Left calcarine -> CTT2	-30.646			-30.646
IQ-> CTT2	-1.26	*		-1.26
Wechsler Memory Scale- Immediate recall (WMS_Im)
Chemo or not -> Right lingual	0.144	*		0.144
Chemo or not -> Left pallidum	-0.019			-0.019
Chemo or not -> Left occipital sup	0.122	**		0.122
Chemo or not -> WMS_Im	0.451		-0.668	-0.217
Right Lingual -> WMS_Im	-4.711			-4.711
Left pallidum -> WMS_Im	23.892	**		23.892
Left occipital sup -> WMS_Im	3.823			3.823
Menopause -> WMS_Im	1.602			1.602
WMS- Delayed recall (WMS_De)
Model 1
Chemo or not -> Left Occipital sup	0.152	*	-0.004	0.148
Fatigue -> Left occipital sup	-0.010			-0.010
Chemo -> Fatigue	0.396			0.396
Chemo or not -> WMS_De	0.050		0.169	0.219
Left occipital sup -> WMS_De	0.985			0.985
Fatigue -> WMS_De	0.060		-0.01	0.049
Menopause -> WMS_De	1.278		-0.055	1.223
Model 2
Chemo or not -> Left Occipital sup	0.149	*		0.149
Menopause -> left occipital	-0.058			-0.058
Left occipital sup-> fatigue	-0.668		0.119	-1.278
Menopause -> fatigue	0.142		0.075	0.217
Chemo -> Fatigue	0.488		-0.19	0.298
Chemo or not -> WMS_De	0.050		0.164	0.214
Left occipital sup -> WMS_De	0.985		-0.076	0.909
Fatigue -> WMS_De	0.060			0.06
Menopause -> WMS_De	1.278		-0.045	1.233
WMS- Recognition (WMS_Re)
Model 1
Chemo or not -> right calcarine	0.24	*******	-0.003	0.237
Fatigue -> Right calcarine	-0.008			-0.008
Menopause -> right calcarine	-0.097			-0.097
Chemo or not -> Left occipital sup	0.152	**	-0.004	0.148
Fatigue-> Left superior occipital	-0.010			-0.01
Menopause-> Left superior occipital	-0.056			-0.056
Chemo or not -> Right thalamus	0.151	**	-0.001	0.150
Fatigue -> Right thalamus	-0.002			-0.002
Menopause -> Right thalamus	-0.052			0.067
Chemo -> fatigue	0.396			0.396
Chemo or not -> WMS_Re	-1.883		0.652	-1.231
Right calcarine -> WMS_Re	-2.973			-2.973
Left occipital sup -> WMS_Re	9.362			9.362
Right thalamus -> WMS_Re	-0.564			-0.564
Fatigue -> WMS_Re	0.146		-0.074	0.072
Menopause -> WMS_Re	3.46	*	-0.209	3.251
Model 2
Chemo or not -> right calcarine	0.237	*******		0.237
Right calcarine -> fatigue	-0.413			-0.413
Menopause -> right calcarine	-0.098			-0.098
Chemo or not -> Left occipital sup	0.149	**		0.149
Left occipital sup-> Fatigue	-1.247			-1.247
Menopause-> Left superior occipital	-0.058			-0.058
Chemo or not -> Right thalamus	0.15	**		0.15
Right thalamus -> fatigue	0.434			0.434
Menopause -> Right thalamus	-0.052			-0.052
Chemo -> fatigue	0.573		-0.218	0.355
Menopause -> fatigue	0.091		0.091	0.091
Chemo or not -> WMS_Re	-1.883		0.654	-1.230
Right calcarine -> WMS_Re	-2.97		-0.06	-3.034
Left occipital sup -> WMS_Re	9.362	**	-0.183	9.18
Right thalamus -> WMS_re	-0.564		0.064	-0.501
Fatigue -> WMS_re	0.146			0.146
Menopause -> WMS_Re	3.46	*	-0.211	3.248
Prospective memory – Event based
Chemo or not -> Left calcarine	0.302	*******		0.302
Chemo or not-> Event based	0.088		0.002	0.091
Left calcarine -> Event based	0.008			0.008
Prospective memory – Time based
Chemo or not -> Time based	-0.298			-0.298
Menopause -> Time based	-0.365			-0.365
IQ-> Time based	0.008			0.008
Models included cancer status, IQ, anxiety, depression, fatigue, and brain areas that were significantly different among the three groups in Tables 1&2. Model 1: status of cancer or chemotherapy as the independent variable, each neurocognitive test as the dependent variable, mood symptoms as the first mediating variable, and ReHo as the second mediating variable; Model 2: same independent and dependent variables, with ReHo as the first mediating variable and mood symptoms as the second mediating variable. Bold letters: the effect was statistically significant: p<0.05, p<0.01, **p<0.001.

No competing interests reported.

Download PDF

Version 1

posted

You are reading this latest preprint version

Associations of cognition, mood symptoms, and brain regional homogeneity in patients with breast cancer with or without chemotherapy and healthy controls

Status:

Version 1

Abstract

Figures

Key Points

Introduction

Methods

Participants

Study procedure

Neuropsychological measures

Functional MRI acquisition

Functional image preprocessing

Reginal homogeneity

Statistical analyses

Results

Discussion

Strengths And Limitations

Conclusion

Declarations

References

Tables

Additional Declarations

Status:

Version 1