Chemotherapy Induced Deficits in Cognition and Affective Behavior

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of Philosophy in the Graduate School of The Ohio State University

Brant Lee Jarrett

Graduate Program in Neuroscience

The Ohio State University

2013

Dissertation Committee:

Anne Courtney DeVries, Advisor

Randy J. Nelson

Jonathan P. Godbout

Georgia A. Bishop

Brant Lee Jarrett

2013

Abstract

Advances in cancer detection and treatment have substantially increased the survival rate for breast cancer patients. Chemotherapeutic drugs, particularly anthracyclines, have potentially toxic side effects in the brain- impacting memory, processing time and verbal fluency, depression, and anxiety. More than 30% of patients undergoing chemotherapy for breast cancer suffer from cognitive deficits, depression, or anxiety during and after treatment. Furthermore, for some women these symptoms persist for decades after treatment cessation. Therefore, alleviating anxiety, depression, and cognitive deficits among cancer survivors is important for both quality of life (QOL) and long-term physical health. However, despite the high incidence and significant impact on quality of life, little is known about the cause of the chemotherapy related effects on cognition and affective behavior and no effective treatment currently exists. The present body of work examined the effects of chemotherapeutic agents doxorubicin and cyclophosphamide (AC) on behavior in a non-tumor bearing mouse model. The goals of this dissertation are to determine the physiological mechanism through which AC impairs cognition and affective behavior, as well as test the efficacy of minocycline, an antibiotic with anti-inflammatory properties, in amelioration of chemotherapy-induced changes in behavior. Chemotherapy treated mice exhibited greater anxiogenic-like and depressive-like behavior than the vehicle group; however, there was not a difference in overall activity level, suggesting that these differences were not due to chemotherapy- induced lethargy. In addition, microglia in the brains of chemotherapy-treated mice had

ii shifted to a proinflammatory state compared to the vehicle group. There was an overall increase in proinflammatory cytokine expression, specifically tumor necrosis alpha

(TNF-α) and interleukin 6 (IL-6), in the chemotherapy-treated mice. We also show that administration of minocycline during chemotherapy reduces expression of IL-6 and prevents depression and anxiety behavior in chemotherapy treated mice.

Chemotherapy also affected learning and memory; both the chemotherapy- treated and vehicle-treated mice were able to acquire the task by the final day of training. However, the vehicle-treated mice reached asymptotic performance after fewer training sessions than the chemotherapy-treated mice. Furthermore, the chemotherapy- treated mice took significantly longer to find the escape box than the vehicle-treated mice on training days 2-5. In contrast, there were no group differences in overall locomotor activity level. In addition, chemotherapy increased proinflammatory cytokine expression (TNF-α, IL-1β, and IL-6) in the hippocampus, a region of the brain that is critical for spatial learning and memory. Minocycline reversed the chemotherapy- induced cognitive deficits and increase in IL-6.

Together, these data suggest that the administration of doxorubicin and cyclophosphamide in doses that are at the lower limit of doses used in women being treated for breast cancer, produce neuroinflammation, including microglia activation, increased proinflammatory cytokine expression and concomitant cognitive deficits and alterations in affective behavior in female mice. Treatment with minocycline reverses the behavioral and inflammatory effects of chemotherapy. These findings suggest a possible mechanism by which chemotherapy alters neuropsychological behavior. More

iii importantly administration of minocycline to cancer patients receiving chemotherapy may improve quality of life and overall physical and mental health; currently no treatments exist for chemotherapy-induced cognitive deficits. Furthermore, minocycline is readily administered in hospitals, is well tolerated over long periods of time, and would be an inexpensive treatment as it is also available in generic form.

Dedication

This thesis is dedicated to my wife, Olivia Olsen Jarrett for being next to me and supporting me every step of the way and to Von and Elaine Jarrett, for guiding and

encouraging me.

Acknowledgments

As a masters student at Brigham Young University studying the effects of stress in humans I became familiar with Courtney DeVries research and wanted to work with animal models in her lab. Over the years Courtney has been influential in instructing and helping me to become a scientist. She was always open to new ideas and new projects and her knowledge of animal behavior was essential in helping me design and carry out successful behavioral studies. I will forever be indebted to her for helping me obtain my PhD.

Kate Karelina and Katie Stuller were also influential in helping me transition from clinical research to animal studies. Kate was very patient in answering all of my questions and Katie broadened my horizon by teaching me about immunology and how to apply it to the various projects I was working on. Adam Hinzey assisted me in a number of different projects and was very helpful in designing and carrying out projects.

Monica was always willing to help with every aspect of my projects and made the lab a fun environment and Zhang Ning truly is an expert surgeon, without whom I would not have been able to work on a number of projects. I owe a great deal of thanks to the

Nelson lab for allowing me to use their equipment and specifically to James Walton who was always willing to show me new techniques and help me modify experimental designs to make them better. Thanks to Chen Shan for performing some of the behavior and other aspects of the experiments and Tracy, Laura, John, and Taryn for all of their insight.

I have had a lot of help from a number of faculty members and collaborators at

OSU. I want to thank Dr. Randy Nelson, Dr. Georgia Bishop, and Dr. Jonathan Godbout for serving on my candidacy and dissertation committees. Your time and input have been invaluable in my success. I want to thank Dr. Maryam Lustberg and Dr. Charles

Shapiro for guidance on the drug administration and making possible this amazing project. I also want to thank Eric Wohleb, Ashley Fenn, Dr. Jon Godbout, and Dr. Kevin

Foust for all of their help and insight with this chemotherapy project specifically. They have been wonderful to collaborate with. I also want to thank all of the undergraduates that spent many hours assisting with all of the tedious aspects involved in research, specifically Christine Fung, Julie Corbett, Mark Serpico, David Paternite, Allison Butler,

Mandy Fritsch, and Andrew Tarr.

Finally I have to thank my family for everything. My wife was willing to follow me out here, away from all of our family and friends, to be a single mom for four and a half years while I continued my education. My parents always encouraged me and believed in me and always pushed me to obtain as much education as possible. My father is my hero and was the perfect example of hard work. Many times as a young teenager I would have to work with him instead of playing with my friends. Although I hated it then, it helped me learn the importance of hard work without which, I wouldn’t be where I am today.

vii

Vita

May 1997 ...... Mountain View High School

2006………………………………….B.S. Psychology, Brigham Young University

2009………………………………….M.S. Psychology, Brigham Young University

2009 to present ...... Graduate Student, Department of

Neuroscience, The Ohio State University

Publications

Stuller, K.A., Jarrett, B., and DeVries, A.C. (2012). Stress and social isolation increase vulnerability to stroke. Experimental Neurology, 233(1):33-9. PMID: 21281636.

Karelina K, Stuller KA, Jarrett B, Zhang N, Well J, Norman G, DeVries AC (2011). Oxytocin mediates social neuroprotection after cerebral ischemia. Stroke, 42 (12), 3606-3611. PMID: 21960564.

Jarrett B, Bloch GJ, Bennett D, Bleazard B, & Hedges D (2010). The influence of body mass index, age and gender on current illness: a cross-sectional study. International Journal of Obesity, 34, 429-436. PMID: 2001903

Whitney T, Hedges D, Brown B, Jarrett B (2008). A comparison between Asperger's syndrome and autism according to clinical and demographic factors in children, adolescents, and young adults. Journal of Therapeutic Schools & Programs, 3, 1, 98-115.

Field of Study

viii

Major Field: Neuroscience

Table of Contents

Abstract ...... ii

Dedication ...... viii

Acknowledgments ...... vi

Vita ...... viiiiii

List of Tables ...... xi

List of Figures ...... xii

Chapter 1: Introduction ...... 1

Chapter 2: Methods ...... 86

Chapter 3: Chemotherapy-induced Changes in Affective Behavior ...... 97

Chapter 4: Chemotherapy-induced Cognitive Deficits ...... 132

Chapter 5: Conclusion ...... 147

References ...... 155

List of Tables

Table 1. Self Report of Cognitive Deficits ...... 53

Table 2. Composite Scores for Global Cognitive Function ...... 55

Table 3. Motor Function ...... 57

Table 4. Processing Speed ...... 58

Table 5. Visuospatial Ability and Visual Learning and Memory ...... 61

Table 6. Working Memory and Executive Function ...... 63

Table 7. Verbal Ability, Learning, and Memory ...... 67

Table 8. Effects of Chemotherapy on Affect in BC Patients ...... 70

Table 9. Effects of Chemotherapy on Cognition in Rodents ...... 72

Table 10. Effects of chemotherapy on Affective Behavior in Rodents ...... 74

Table 11. Chemotherapy-induced Inflammatory Changes in Clinical Studies ... 75

Table 12. Chemotherapy-induced Changes in Brain Structure ...... 77

Table 13. Pre-clinical Mechanistic Studies ...... 80

List of Figures

Figure 3.1. Chemothearpy (72 hours) Increases Protein Expression of TNFα and CCL4 in the Brain ...... 115

Figure 3.2. Chemotherapy Increases mRNA Expression of TNFα and IL-6 .... 116

Figure 3.3. Chemotherapy Increase Time Spent Floating in the Forced Swim Test

...... 117

Figure 3.4. Chemotherapy Increases Time Spent in EPM Open Arms ...... 118

Figure 3.5. Chemotherapy Does Not Affect Locomotor Activity in BALB/C Mice119

Figure 3.6. Chemotherapy Increases Time Spent Floating in C57bl/6 Mice..... 120

Figure 3.7. Chemotherapy Increases Time Spent in EPM Open Arms in C57bl/6 Mice

...... 121

Figure 3.8. Chemotherapy Does Not Affect Locomotor Activity in C57bl/6 Mice122

Figure 3.9. Chemotherapy Increases Microglial Activation Markers in the CNS123

Figure 3.10. Concentrations of AC and Phosphamide Mustard in the Blood and Brain

...... 124

Figure 3.11. Minocycline (i.c.v.) Ammeliorates Chemotherapy-induced Immobility in the

Forced Swim Test ...... 125

Figure 3.12. Minocycline (i.c.v.) Ammeliorates Chemotherapy-induced Anxiety-like

Behavior in the EPM...... 126

xii

Figure 3.13. Minocycline (i.c.v.) Does Not Affect Overall Locomotor Activity ... 127

Figure 3.14. Minocycline Ameliorates Chemotherapy-induced Gene Expression of Pro- inflammatory Cytokine IL-6 in the Prefrontal Cortex ...... 128

Figure 3.15. Minocycline (oral) Ammeliorates Chemotherapy-induced Immobility in the

Forced Swim Test ...... 129

Figure 3.16. Minocycline (oral) Does Not Affect Overall Locomotor Activity ..... 130

Figure 3.17. Minocycline Ameliorates Chemothearpy-induced Increases in Gene

Expression of IL-6 ...... 131

Figure 4.1. Chemotherapy Impairs Learning and Memory in the Barnes Maze 141

Figure 4.2. Chemotherapy Increases mRNA Expression of IL-1β and IL-6 ..... 142

Figure 4.3. Chemotherapy Increases mRNA Expression of TNFα ...... 143

Figure 4.4. Chemotherapy Decreases Dendritic Spine Density in the Hippocampus

...... 144

Figure 4.5. Monicycline Ameliorates Chemotherapy-induced Cognitive Impairment

...... 145

Figure 4.6. Monicycline Ameliorates Chemotherapy-induced Gene Expression of Pro- inflammatory Cytokine IL-6 in the Prefrontal Cortex ...... 146

xiii

Chapter 1: Introduction

1. Background 2. Cognitive effects in breast cancer patients 3. Depression and anxiety in breast cancer patients 4. Evidence of Chemotherapy-related Changes in Brain Structure and Function 5. Neuropsychological effects in rodents 6. Mechanism 6.1 Changes in cytokines in humans 6.2 Pre-clinical studies 7. Roadblocks 7.1 Rodent models 7.2 Drug administration 8. Conclusion

Background

Breast cancer is the most frequently diagnosed cancer in women; one in eight women in the U.S. will be diagnosed with this disease during their lifetime

(Howlader N, 2011). Due to improved detection and treatment advances, the overall 5-year and 10-year survival rates for breast cancer (BC) are now 90% and 84.5% respectively (Howlader N, 2011). Thus, there is a growing population of breast cancer survivors for whom persistent side effects of treatment could impinge upon quality of life. The symptoms most commonly reported to reduce quality of life among breast cancer patients include cognitive deficits, depression, 1 anxiety, fatigue, and sleep disturbance (Bower et al., 2000; Burgess et al., 2005; de Jong, Courtens, Abu-Saad, & Schouten, 2002; Fiorentino & Ancoli-Israel,

2006; Myers, 2012). While most studies focused on these symptoms individually, they often occur in clusters (Fiorentino, Rissling, Liu, & Ancoli-Israel, 2012; Ganz et al., 2013; Liu et al., 2009; Rutledge, 2003; So et al., 2009). In a longitudinal study assessing the effects of chemotherapy on fatigue, depression, and sleep disturbance, all symptoms were moderately to strongly correlated from baseline

(before chemotherapy) until three weeks after the final treatment, and all women reported impaired sleep, fatigue, and depressive symptoms during treatment compared to baseline (Liu et al., 2009). Furthermore, fatigue, depression, and cognitive impairment worsened across time. Given that fatigue and sleep disruption are among the diagnostic criteria for depression (Fann et al., 2008), it is not surprising that depression is a better predictor of quality of life than either fatigue or insomnia alone(Redeker, Lev, & Ruggiero, 2000). Thus, a larger percentage of the women being treated for BC using chemotherapy develop neuropsychological symptoms that can have a drastic impact on everyday life; the specific focus of this review will be chemotherapy-related deficits on cognitive function and affect.

Chemotherapy-related cognitive impairment (CRCI), often informally referred to as “chemobrain” or “chemofog”, is commonly reported by breast cancer patients (Myers, 2012). The majority of studies with BC patients that incorporate cognitive self-report measures indicate that women who have been

2 treated with chemotherapy report greater cognitive deficits than BC patients who have not been treated with chemotherapy and matched healthy controls (Table

1). However, the self-reports rarely correlate with objective measures of cognitive deficits. Indeed, results from clinical studies have been largely inconsistent (see Tables 1-7); the effects of chemotherapy on cognition, the percentage of patients affected, and the duration of the cognitive deficits vary greatly across studies. Both individual and experimental factors are likely to contribute to the disparate conclusions among the studies.

In addition to cognitive dysfunction, BC patients often report changes in mood. Depression has received more attention among cancer patients than any other psychiatric disorder (Caplette-Gingras & Savard, 2008) and incidence may be as high as 41% among breast cancer patients (So et al., 2010) . Anxiety has received less attention, although it is frequently associated with depression; approximately 45% of breast cancer patients have moderate to high levels of anxiety (Bjorneklett et al., 2012). Depressive symptoms are associated with decreased survival, increased recurrence rates, tumor growth, and decreased quality of life in breast cancer patients, although it is not clear if a causal relationship exists between depression and these outcome measures (Falagas et al., 2007; Fann et al., 2008; Goodwin, Zhang, & Ostir, 2004; Khan, Amatya,

Pallant, & Rajapaksa, 2012; Redeker, Lev, & Ruggiero, 2000). However, the effects of chemotherapy on depression and anxiety are understudied and prevalence rates are inconsistent across studies. The inconsistency may be due

3 to the frequency of self-report measures used to assess these disorders and difficulty in making a clinical diagnosis of depression in cancer patients with physical symptoms; some of the physical side effects of chemotherapy and cancer overlap with DSM-IV criteria for depression, namely lack of energy, poor concentration, and sleep disturbance (Fann et al., 2008).

While there are many treatment factors and individual susceptibilities that may increase the risk of cognitive and affective deficits among BC patients, cognitive decline and depression emerge most consistently and may be the most debilitating among women whose treatment plan included chemotherapy

(Redeker, Lev, & Ruggiero, 2000; Stewart, Bielajew, Collins, Parkinson, &

Tomiak, 2006). However, the mechanisms through which chemotherapy may be affecting brain function remains a matter of speculation; to date, no causal mechanism has been established in either humans or other animals.

Furthermore, there are currently no standard pharmacological treatments for chemotherapy-induced cognitive dysfunction and treatment of depression using conventional pharmaceutical methods is limited by potential interference with the effectiveness of endocrine therapy (ET; Kelly et al., 2010). In sum, the neuropsychological side effects of chemotherapy tend to be under-recognized and undertreated, which in turn can compromise quality of life by amplifying the physical symptoms and functional impairments experienced by the women (Fann et al., 2008). In addition, untreated depression and concern about side effects can contribute to poor treatment adherence (Colleoni et al., 2000).

The goals of this review are: 1) to provide a synthesis of the chemotherapy-related side effects on cognition and mood/affective-like behaviors in both human and animal models, 2) provide an overview of the possible data supporting different mechanistic hypotheses, and 3) identify the barriers that have hampered translational research into the causes and potential treatments for the neuropsychological side-effects of chemotherapy.

2. Cognitive effects of chemotherapy in breast cancer patients

Nearly forty years ago the possibility of antineoplastic drugs having neurotoxic side-effects was first reported (Weiss, Walker, & Wiernik, 1974). In

1980, researchers provided evidence in support of this hypothesis by reporting that non-CNS cancer patients being treated with chemotherapy were experiencing cognitive deficits (Silberfarb, Philibert, & Levine, 1980). Since then, the effects of chemotherapy on cognition have garnered much interest and more than a thousand papers on this topic have been published, many in breast cancer patients due to the relatively high survival rate. However, the mechanism by which CRCI occurs remains unknown. Although there is a vast literature discussing the cognitive side effects of chemotherapy in breast cancer patients, the study designs vary drastically, making it difficult to generate cross-study comparisons (Nelson & Suls, 2013; Shilling, Jenkins, & Trapala, 2006; Vardy,

Rourke, & Tannock, 2007; Wefel, Vardy, Ahles, & Schagen, 2011). As seen in

Tables 1-7, there are many differences across studies, including chemotherapy

5 agents used, subject age, cognitive instruments, menopause status, treatment duration, concomitant endocrine therapy (ET), and time of testing. Other variations also include race, inclusion criteria, and other treatments. Even within each study there are often differences in chemotherapy agents used, stage of disease, and treatment schedule.

One major issue is the difference in tests used to assess cognitive ability.

The studies compiled in Tables 1-7 used more than thirty different cognitive tests to assess cognitive functioning. These assessments are generally broken down into the domains of visuospatial ability, learning, language, attention, working memory, concentration, reaction time, psychomotor speed, and executive functioning. Furthermore, different methods of calculating cognitive scores are used to determine cognitive ability (Shilling, Jenkins, & Trapala, 2006). Shilling

(2006) demonstrated the importance of having an appropriate analytical plan by analyzing cognitive impairment in one cohort of breast cancer patients using seven different methods that had been reported in the literature; depending on the method of analysis, cognitive impairment ranged from 12% to 68% in chemotherapy treated patients and 4% to 64% in the healthy controls. In an attempt to encourage standardization across studies and to facilitate consideration of the multitude of individual and treatment factors that can contribute to cognitive deficits in cancer patients, the International Cognition and

Cancer Task Force recommended a core set of neuropsychological tests and common criteria for assessing cognitive impairment in cancer patients (Wefel,

Vardy, Ahles, & Schagen, 2011). They recommended prospective, longitudinal trials, including disease specific and healthy control groups (local controls and published normative data), with cognitive assessment at baseline (pretreatment), and repeat assessment whenever possible to assess changes in cognitive function over time. They also recommend assessing the cognitive domains of learning and memory, processing speed, and executive functioning, using Hopkins Verbal Learning Test-Revised, Trail Making Test, and the

Controlled Oral Word Association (COWA) of the Multilingual Aphasia

Examination, respectively. These tests were recommended because they assess commonly occurring cancer-related impairments, have adequate psychometric properties, have alternate forms to limit practice effects, and have been translated into multiple languages (Wefel, Vardy, Ahles, & Schagen, 2011).

Study design also has been a major issue in assessing cognitive deficits in

BC patients. Until recently, most studies were cross-sectional rather than longitudinal. In fact, meta-analyses published in 2005 and 2006 (Falleti,

Sanfilippo, Maruff, Weih, & Phillips, 2005; Stewart, Bielajew, Collins, Parkinson,

& Tomiak, 2006) on the cognitive effects of chemotherapy included only one longitudinal study among 11 cross-sectional studies. While the number of longitudinal studies of cognitive deficits in BC patients has increased recently, they still comprised less than half of those incorporated in a recent meta-analysis

(8 of 17; Jim et al., 2012) Jim 2012). The advantage of the early cross-sectional studies is that they could be completed in a shorter period of time and at lower

7 cost than longitudinal studies, while providing an indication of the prevalence of cognitive deficits among BC patients. An important disadvantage is that cross- sectional design cannot rule out that cognitive deficits existed among the patients prior to chemotherapy, nor do they provide information about how cognitive function changes within an individual in response to treatment and with the passage of time; indeed, one early cross-sectional study reported that 35% of BC patients exhibited cognitive deficits prior to chemo treatment when compared to normative data (Wefel et al., 2004). Importantly, however, the existence of cognitive deficits prior to systemic treatment, does not rule out a potential additive or synergistic effect of chemotherapy treatment on cognitive dysfunction.

Several subsequent cross-sectional studies have attempted to address this issue by including patients who received chemotherapy versus radiotherapy, endocrine therapy or no additional treatment. Assessing cognitive function six months after completion of treatment resulted in no cognitive differences between breast cancer patients treated with both radiotherapy and chemotherapy and those only receiving radiotherapy (Donovan et al., 2005). Findings from patients assessed an average of 2 years after treatment did report, however, greater cognitive impairment in breast cancer patients treated with chemotherapy relative to those only receiving radiotherapy (Schagen et al., 1999). Results from studies assessing cognitive functioning five years after treatment with chemotherapy- versus radiotherapy-also reported conflicting results, one finding chemotherapy related cognitive impairments (Ahles et al., 2002), and one reporting no

8 differences (Scherwath et al., 2006). These findings reiterate the importance of performing longitudinal studies, by which cognitive functioning can be assessed before and after chemotherapy treatment for each patient.

In the past decade, numerous longitudinal studies have been published, thus clarifying that many BC patients experience a chemotherapy-related cognitive decline. The vast majority of these longitudinal studies report that breast cancer patients treated with chemotherapy experience a significant decrease in cognitive ability compared to non-chemotherapy treated patient and healthy controls. Findings from three longitudinal studies reported deficits in

11%-31% of patients prior to initiation of chemotherapy (Hermelink et al., 2007;

Hurria et al., 2006; Jansen, Dodd, Miaskowski, Dowling, & Kramer, 2008). Two of these three studies, however, reported that even with these deficits at baseline, chemotherapy was associated with a further decline in cognitive function across time (Hurria et al., 2006; Jansen, Dodd, Miaskowski, Dowling, & Kramer, 2008).

The third study(Hermelink et al., 2007) found that 27% of the chemotherapy patients experienced a cognitive decline relative to the pre-chemotherapy baseline while 28% demonstrated improvement relative to baseline over the same time period; one drawback of this latter study, however, is that the last session of cognitive testing occurred before the chemotherapy regimen had been completed. This is problematic as others have reported a significant relationship between the number of chemotherapy treatments and degree of cognitive impairment (Hodgson, Hutchinson, Wilson, & Nettelbeck, 2013). Other

9 longitudinal studies have reported no difference at baseline between chemotherapy treated patients and healthy controls (Bender et al., 2006;

Mehlsen, Pedersen, Jensen, & Zachariae, 2009; Schagen, Muller, Boogerd,

Mellenbergh, & van Dam, 2006). These findings suggest that while other factors related to cancer may have an effect on cognition, chemotherapy treatment itself has a negative impact on cognitive functioning.

Another crucial issue for BC patients is how long cognitive deficits are likely to persist following completion of chemotherapy. The long-term consequences of chemo on cognition have also been well documented in recent years. While a few studies have suggested that cognitive ability is restored within one year of treatment cessation (Collins, Mackenzie, Stewart, Bielajew, & Verma,

2009), most studies report that cognitive deficits can be detected up to 15 years after the final chemotherapy treatment (Ahles et al., 2002; Bender et al., 2006;

Hodgson, Hutchinson, Wilson, & Nettelbeck, 2013; Hurria et al., 2006; Schagen et al., 1999; Shilling, Jenkins, Morris, Deutsch, & Bloomfield, 2005; Wefel, Lenzi,

Theriault, Davis, & Meyers, 2004; Yamada, Denburg, Beglinger, & Schultz,

2010). Indeed, a recent study revealed that BC survivors who were treated with chemotherapy 10-15 years prior to testing demonstrated impaired performance on multiple cognitive tests relative to age matched controls without a prior cancer diagnosis (Yamada, Denburg, Beglinger, & Schultz, 2010). Likewise, a similar cognitive deficit was apparent among breast cancer and lymphoma survivors who had received chemotherapy 5-15 years before testing (Ahles et al., 2002).

These studies were cross-sectional, so there is no information available about changes in cognitive ability that may have occurred during the intervening decade and a half, but several longitudinal studies with shorter recovery periods

(3-18 months) provide added support for persistent cognitive deficits in at least a subset of BC patients (Ahles et al., 2010; Bender et al., 2006; Jansen, Dodd,

Miaskowski, Dowling, & Kramer, 2008; Quesnel, Savard, & Ivers, 2009).

Furthermore, a recent meta-analysis concludes that time since treatment is not associated with improvement in cognitive functioning after chemotherapy

(Hodgson, Hutchinson, Wilson, & Nettelbeck, 2013). The number of chemotherapy treatments, however, was negatively associated with cognitive performance; more treatments results in poorer cognitive functioning (Hodgson,

Hutchinson, Wilson, & Nettelbeck, 2013). This suggests that the duration of chemotherapy treatment may be a better indicator of long term deficits than the time elapsed since completion of chemotherapy. Chemotherapy regimen is also an important factor in determining incidence of cognitive deficits; a high-dose regimen of cyclophosphamide, thiopeta, and carboplatin is nearly four times as likely to result in deterioration in cognitive performance as a standard-dose of 5- fluorouracil, epirubicin, and cyclophosphamide (25% versus 7%, respectively;

Schagen, Muller, Boogerd, Mellenbergh, & van Dam, 2006). Similarly, breast cancer patients receiving adriamycin, cyclophosphamide, and fluorouracil had more cognitive complaints than those receiving cyclophosphamide, methotrexate, and fluorouracil (Janelsins et al., 2012). Thus, the treatment

11 regimen given is also important factors to consider when assessing the effects of chemotherapy on cognitive functioning.

While it is widely accepted that a decline in cognitive functioning is a possible consequence of chemotherapy in at least a subset of breast cancer patients, the wide variations in study design, cognitive tests, chemotherapy drugs, treatment duration and method of data analysis make it difficult to provide evidence for the mechanisms underlying chemotherapy-induced cognitive deficits and to identify the subset of patients who are likely to be at risk. Studies of both adequate sample size and consistency in variables such as chemotherapy regimen, treatment duration, and data analysis will be instrumental in future research.

3. Depression and anxiety in breast cancer patients

The prevalence of depression in cancer patients has been well studied and ranges from 5% to 50% among breast cancer patients (Caplette-Gingras &

Savard, 2008; Knobf, 2007; Massie, 2004; Reece, Chan, Herbert, Gralow, &

Fann, 2013). Depression is associated with impairment in quality of life and has a negative influence on compliance of medical treatment, cancer recurrence, recovery, and survival (Colleoni et al., 2000; Falagas et al., 2007; Goodwin,

Zhang, & Ostir, 2004; Khan, Amatya, Pallant, & Rajapaksa, 2012; Redeker, Lev,

& Ruggiero, 2000). Factors associated with increased affective symptoms among

BC patients include younger age, fatigue, pain, insomnia, metastasis, and lack of

12 social support (Ahles et al., 2002; Avis et al., 2012; Caplette-Gingras & Savard,

2008; Greenberg, 2004; Hopwood, Howell, & Maguire, 1991; Hotopf, Chidgey,

Addington-Hall, & Ly, 2002; Kissane et al., 2004; Koopman et al., 2002).

Comorbidity of depression and anxiety is common in primary care, with up to 67% of those with a depressive disorder having a comorbid anxiety disorder and 63% with an anxiety disorder also having a current depressive disorder

(Hirschfeld, 2001; Lamers et al., 2011). Anxiety is less frequently studied than and often combined with depression as the two are frequently associated. For example, one study reported that nearly 50% of women experienced an episode of depression, anxiety, or both within one year of being diagnosed with breast cancer without reporting the prevalence for depression or anxiety alone (Burgess

2005). Studies that have reported anxiety levels in breast cancer patients show up to 48% having high levels of anxiety within 3 months of being diagnosed

(Boyes et al., 2013; Vahdaninia, Omidvari, & Montazeri, 2010). Anxiety itself is associated with cognitive decline and reduced quality of life (So et al., 2010;

Vearncombe et al., 2009). Thus, it is important to assess the effects of anxiety separate from those of depression as well as together.

Relatively few studies have focused specifically on the effects of chemotherapy on depression and anxiety in breast cancer patients, most of them being published within the past 5 years. The prevalence rates vary greatly across studies as there are differences in chemotherapy regimen, treatment duration, testing used and determinant of affective disorder (i.e. cutoff score or DSM-IV

13 diagnosis). Furthermore, several studies fail to report the specific chemotherapy regimen or separately analyze the drugs to determine if some regimens are more commonly associated with affective disorders (see Table 8). These discrepancies make it difficult to compare across studies; indeed, one group of researchers recently published a systematic review on anxiety in breast cancer patients after concluding that the methodological heterogeneity across studies was too great to support a meta-analysis (Lim, Devi, & Ang, 2011).

As seen in Table 8, not only are there a number of different chemotherapy agents used across studies and within studies, but a number of studies don’t mention the agents used. This is problematic as the specific chemotherapy drug administered may be an important factor influencing the precipitation of depression and anxiety. For example, a recent cross-sectional study reports that treatment with a doxorubicin-based regimen is nearly three times as likely to result in depressive symptoms as a regimen not containing doxorubicin or a non- chemotherapy treatment (Avis et al., 2012). Similarly, a longitudinal study found an increase in depression when treated with doxorubicin/cyclophosphamide, but not other chemotherapy regimens (Reece, Chan, Herbert, Gralow, & Fann,

2013).

While most results suggest that chemotherapy is associated with increased depression and anxiety (Bower et al., 2011; Hwang, Chang, & Park,

2013; Reece, Chan, Herbert, Gralow, & Fann, 2013; Torres et al., 2013), drawing conclusions about the duration of this effect is difficult to do because there are

14 very few longitudinal studies that extend beyond one year after treatment and include a BC non-chemotherapy control group; a vast majority assessed the effects of an psychological intervention program on affective behavior in patients who all received chemotherapy (Bjorneklett et al., 2012; Kissane et al., 2003;

Kissane et al., 2004; Komatsu et al., 2012). However, a recent longitudinal study found that even three years after treatment cessation, chemotherapy treated breast cancer patients had more depressive symptoms than non-chemotherapy treated patients (Hwang, Chang, & Park, 2013). This is in contrast to findings that depression and anxiety generally improve within one year of being diagnosed with breast cancer (Collins, Mackenzie, Stewart, Bielajew, & Verma, 2009; Wefel,

Lenzi, Theriault, Davis, & Meyers, 2004); these data suggest that chemotherapy can induce long-lasting effects on depression and anxiety. More long-term longitudinal studies are needed to determine the possible long-term side effects of chemotherapy-related affective disorders on quality of life, recurrence of cancer, and mortality rates. Furthermore, while some patients may improve, not all patients do and how marginal or significant the improvement is yet to be determined. It is also imperative to determine the risk factors for those who do not improve and determine whether chemotherapy treatment is inducing persistent changes in the brain that may precipitate or sustain affective disorders in BC patients.

As briefly discussed above, in the past few years, researchers have found that a number of cancer and chemotherapy-related side effects associated with a

15 decrease in quality of life frequently occur together in cancer patients. This co- occurrence of symptoms is known as a symptom cluster (Fiorentino, Rissling,

Liu, & Ancoli-Israel, 2012). The symptoms most commonly co-occurring are depression, anxiety, fatigue, pain, and trouble sleeping/insomnia (Fiorentino,

Rissling, Liu, & Ancoli-Israel, 2012; Liu et al., 2009; Sanford et al., 2013; So et al., 2009). Depression and anxiety are frequently co-morbid in cancer patients

(van den Beuken-van Everdingen et al., 2009) and some of the physical side effects of chemotherapy and cancer overlap with DSM-IV criteria for depression, namely lack of energy, poor concentration, and sleep disturbance (Fann et al.,

2008). Thus, it is not surprising that these symptoms “cluster” together and that depression is a better predictor of quality of life than fatigue or insomnia alone

(Redeker, Lev, & Ruggiero, 2000). However, a recent study found that while depressive symptoms, sleep problems, and fatigue were positively correlated with each other in breast cancer patients, only fatigue was correlated with inflammatory marker tumor necrosis factor receptor two (TNFRII), suggesting that these symptoms may have different underlying biologic mechanisms (Bower et al., 2011).

While some studies suggest that chemotherapy is associated with depression and anxiety, as well as fatigue, pain, and sleep disturbance, in at least a subset of breast cancer patients, the wide variations in study design, affective tests, chemotherapy drugs, and failure to report methodologies makes it difficult to form hypotheses regarding the mechanism of chemotherapy-induced

16 changes in affect. Longitudinal studies reporting essential study parameters and maintaining consistency in variables such as chemotherapy regimen and treatment duration will be instrumental in future research.

4. Evidence of Chemotherapy-related Changes in Brain Structure and Function

The incorporation of neuroimaging techniques has been instrumental in assessing a neural basis for chemotherapy-related neuropsychological effects.

Both structural and functional studies have been performed, using magnetic resonance imaging (MRI) to assess structural differences, and functional MRI

(fMRI), positron emission tomography (PET), and electroencephalography (EEG) to assess functional differences.

Structural imaging uses various MRI techniques to determine tissue density or volume differences for the whole brain or a specific region of interest.

The first study in a breast cancer population reported that compared to healthy controls, chemotherapy treated patients had increased white matter lesions

(Brown et al., 1995). A follow-up, longitudinal study found that these lesions occur shortly after the initiation of chemotherapy, persisting through the one year follow up assessment (Brown et al., 1998). Only four patients completed all assessments and a pre-chemotherapy assessment was not performed; however, these early findings suggested that chemotherapy is associated with structural changes within the brain, highlighting the need for additional research. A similar study, using a larger sample size reported that one year after treatment,

17 chemotherapy patients had smaller volumes of both gray and white matter in areas related to cognitive processing relative to non-chemotherapy controls

(Inagaki et al., 2007); furthermore, volume loss was positively correlated with cognitive deficits in attention and memory. Since then, a number of cross- sectional and longitudinal MRI studies have reported that chemotherapy is associated with changes in gray matter and white matter in various regions of the brain, as well as total brain volume (See table 6; Abraham et al., 2008; de Ruiter et al., 2011; Deprez et al., 2012; Deprez et al., 2011; S. Kesler et al., 2013;

Koppelmans et al., 2012; McDonald, Conroy, Ahles, West, & Saykin, 2010).

Furthermore, these changes in brain structure are correlated with deficits in attention (Deprez et al., 2011), verbal memory (S. Kesler et al., 2013), executive functioning (S. R. Kesler et al., 2013) and processing speed (Abraham et al.,

2008). The imaging studies suggest that these chemotherapeutic-induced changes are widespread rather than localized to a specific brain region, as there were diffuse changes of gray and white matter (Deprez et al., 2012; Koppelmans et al., 2012; Silverman et al., 2007). These changes can be long lasting, as cross-sectional studies reveal that chemotherapy-related changes are detected in patients up to 21 years after cessation of chemotherapy treatment (de Ruiter et al., 2011; S. Kesler et al., 2013; Koppelmans et al., 2012).

Functional imaging is used to measure brain activity, both global and regional. Functional magnetic resonance imaging (fMRI) is a non-invasive technique used to measure brain activity by detecting oxygen changes in blood

18 volume, the blood oxygen level dependent (BOLD) signal showing regions of high oxygen uptake. BOLD activation in the prefrontal cortex was lower in chemotherapy treated patients than healthy controls during a memory encoding test three years after treatment cessation (Kesler, Bennett, Mahaffey, & Spiegel,

2009). However, BOLD activation was higher in the chemotherapy group across multiple brain regions during a memory recall task; this suggests that chemotherapy may be associated with memory encoding-related functional deficits, and that recalling information is more difficult, thus requiring a greater recruitment of brain regions to perform the given task. Comparing chemotherapy treated patients to non-chemotherapy and healthy controls, findings suggest that both breast cancer and chemotherapy are independently associated with functional changes in the brain (Kesler, Kent, & O'Hara, 2011). Breast cancer also was associated with hypoactivation in the left medial prefrontal and premotor cortex, whereas only the chemotherapy treated group showed reduced activation of the left lateral prefrontal cortex. These results corroborate findings from pre- chemotherapy treatment studies showing that cancer patients have higher level of activation of the inferior frontal gyrus (Cimprich et al., 2010; Scherling, Collins,

Mackenzie, Bielajew, & Smith, 2011). These data reiterate the importance of longitudinal studies and non-chemotherapy controls in assessing the effects of chemotherapy on changes in the brain.

Positron emission tomography, a nuclear imaging technique, produces a three-dimensional image by detecting pairs of gamma rays emitted by a tracer

19 introduced to the body. This technique can be used to assess metabolic activity in the brain. To date only one study has used PET in chemotherapy-related breast cancer studies (Silverman et al., 2007). Chemotherapy patients showed increased activation in the frontal cortex, specifically the inferior frontal gyrus

(IFG) and posterior cerebellum, during the short term memory recall task.

Furthermore, patients treated with tamoxifen, an estrogen antagonist, in addition to chemotherapeutic agents showed reduced metabolism in the basal ganglia, suggesting a possible effect of hormonal therapy on brain function.

Electroencephalography records electrical activity within the brain by measuring voltage fluctuations from neuronal ionic currents. Only one study has assessed chemotherapeutic-related changes in electrical current in the brain

(Schagen, Hamburger, Muller, Boogerd, & van Dam, 2001). While there were no differences for most measures, chemotherapy was associated with asymmetry of alpha rhythm as 41% of patients treated with high dose and 12% treated with standard dose showed alpha rhythm asymmetry compared to 0% of non- chemotherapy controls.

In sum, neuroimaging is a powerful technique that has only recently begun to be used to assess the effects of chemotherapeutic agents on brain structure and function. These findings overwhelmingly support the hypothesis that chemotherapy has detrimental effects within the brain and that these effects may be persistent. Interestingly, the imaging studies have been far more consistent in

20 documenting cognitive deficits than the neuropsychological studies that do not incorporate imaging.

5. Neuropsychological effects of chemotherapy in rodents

Taken together, the clinical studies in breast cancer patients provide convincing evidence that a significant proportion of women undergoing chemotherapy develop persistent cognitive and affective side effects.

Furthermore, the imaging studies suggest that brain structure and function is being affected, even though there is very little evidence that the chemotherapy drugs are passing through the blood brain barrier at high concentrations to directly affect brain cells (Brufsky et al., 2011; Grimm, 2011). However, the fact that there are robust effects on complex behaviors and measureable changes in brain morphology that persist well beyond metabolism of the chemotherapy agents requires that brain function is being either directly or indirectly affected by the chemotherapy drugs. Establishing the physiological mechanism through which chemotherapy is altering brain structure and function can only be accomplished through the use of animal models. Furthermore, this approach allows a systematic and controlled study of the physiological changes in the brain associated with treatment with individual or combined chemotherapeutic agents.

Lastly, rodent models allow high throughput testing of potential pharmacological therapies; currently there are no established methods for preventing or treating the cognitive side effects of chemotherapy and some of the most commonly

21 prescribed selective serotonin reuptake inhibitors (SSRIs)/ selective serotonin and norepinephrine reuptake inhibitors (SNRIs) used to treat depression can reduce the effectiveness of tamoxifen in breast cancer patients (Binkhorst et al.,

2013; Kelly et al., 2010; Roscoe et al., 2005). Thus, the study of the cognitive and affective consequences of chemotherapy is at a juncture in which improved understanding of the mechanisms underlying chemotherapy’s effects on brain and behavior could facilitate the development of effective treatments.

While there are a wide range of chemotherapeutic agents used to treat cancer patients, most animal models focus on doxorubicin, cyclophosphamide, 5- fluorouracil, or methotrexate. As seen in Table 9, a majority of research has shown that chemotherapeutic agents do indeed induce deleterious cognitive effects in non-tumor models (ElBeltagy et al., 2010; Konat, Kraszpulski, James,

Zhang, & Abraham, 2008; Y. Li, V. Vijayanathan, M. E. Gulinello, & P. D. Cole,

2010; Mondie, Vandergrift, Wilson, Gulinello, & Weber; Reiriz et al., 2006;

Winocur et al., 2012; M. Yang et al., 2010). A few studies, however, have reported no effects of chemotherapy on cognition (Boyette-Davis & Fuchs, 2009;

Fremouw, Fessler, Ferguson, & Burguete, 2012; Gandal, Ehrlichman, Rudnick, &

Siegel, 2008; Long et al., 2011; Lyons, Elbeltagy, Bennett, & Wigmore, 2011;

Stock, Rosellini, Abrahamsen, McCaffrey, & Ruckdeschel, 1995) while one reported that chemotherapy improved learning (Lee et al., 2006). These contrasting results are likely due to the differences in treatment protocol and tests used for cognitive assessment. For example a single injection of .005mg/kg

22 methotrexate was administered in one study that failed to detect a chemotherapy effect on cognition (Stock, Rosellini, Abrahamsen, McCaffrey, & Ruckdeschel,

1995), whereas most studies used at least 37 mg/kg and administered multiple injections (Y. Li, V. Vijayanathan, M. Gulinello, & P. D. Cole, 2010; Madhyastha,

Somayaji, Rao, Nalini, & Bairy, 2002). Furthermore, three of the studies with negative results used fear conditioning as the cognitive test, which, as discussed later, may not be a good test for assessing chemotherapeutic-related changes on cognitive ability (Fremouw, Fessler, Ferguson, & Burguete, 2012; Gandal,

Ehrlichman, Rudnick, & Siegel, 2008; Long et al., 2011). Another study that found no differences used a cognitive test (5CSRTT) not commonly used in these other studies (Boyette-Davis & Fuchs, 2009); in the discussion section the authors acknowledge that this may not be a proper task for this type of study.

Thus, it is important to individually assess the drugs and tests used.

Doxorubicin and cyclophosphamide are the most commonly used chemotherapy regimen for the treatment of breast cancer. In rodent models, cyclophosphamide impairs passive avoidance task learning, novel object recognition, and inhibitory avoidance in mice (Reiriz et al., 2006; M. Yang et al.,

2010) while doxorubicin impairs inhibitory avoidance in rats (Liedke et al., 2009).

Combined, these drugs impair passive avoidance and contextual fear, but not cued fear in female rats (Konat, Kraszpulski, James, Zhang, & Abraham, 2008;

Macleod et al., 2007). A few studies found no difference in spatial memory

(MWM and NLR), fear conditioning, or novel object recognition (Fremouw,

Fessler, Ferguson, & Burguete, 2012; G. D. Lee et al., 2006; Long et al., 2011;

Lyons, Elbeltagy, Bennett, & Wigmore, 2011). However, one study did not perform the tasks until 2-10 months after the last chemotherapy injection (G. D.

Lee et al., 2006) compared to most studies testing within one week of treatment

(Liedke et al., 2009; Macleod et al., 2007; Reiriz et al., 2006; Winocur et al.,

2012). Two of the studies used strains (F344 rats and C57BL/6 mice) not used in the other studies (Fremouw, Fessler, Ferguson, & Burguete, 2012; Long et al.,

2011); one of the studies acknowledged that susceptibility to the side effects of chemotherapy may vary by strain (Fremouw, Fessler, Ferguson, & Burguete,

2012). It is also interesting to note that four of the five studies that found impairments used female rats or male mice, while three of the four studies showing no differences using the opposite sex, male rats or female mice.

Methotrexate (MTX) and 5-Fluorouracil (5-FU) are the other two drugs most commonly tested in animal models. Although one study found no effect of MTX+

5-FU on contextual fear conditioning or novel object recognition in male mice

(Gandal, Ehrlichman, Rudnick, & Siegel, 2008), multiple studies have reported that rats and female mice treated with MTX (alone or with 5-FU) demonstrate cognitive impairments in spatial and visual memory (MWM and NLR), but not cued memory (Fardell, Vardy, Logge, & Johnston, 2010; Y. Li, V. Vijayanathan,

M. E. Gulinello, & P. D. Cole, 2010; Madhyastha, Somayaji, Rao, Nalini, & Bairy,

2002; Seigers et al., 2009; Winocur, Binns, & Tannock, 2011; Winocur et al.,

2012; Winocur, Vardy, Binns, Kerr, & Tannock, 2006). Furthermore, the study

24 finding no difference used mice surgically implanted with electrodes in the hippocampus, which may have caused baseline alterations in behavior (Gandal,

Ehrlichman, Rudnick, & Siegel, 2008). Only three studies administered 5-FU alone, two of these reported impairments in spatial memory (NLR) and one in fear conditioning (Lyons, ElBeltagy, Bennett, & Wigmore, 2012; Mustafa, Walker,

Bennett, & Wigmore, 2008). The third study (Fremouw, Fessler, Ferguson, &

Burguete, 2012) found no differences in fear conditioning or novel object recognition. A possible explanation may be the use of C57bl/6 female mice, the only such study of those found in table 9, as discussed previously. Future studies need to use a wider range of tests in C57bl/6 mice to address this issue. Three other drugs have only been used once each in rodent models, finding cognitive impairments when treated with Oxaliplatin or thioTEPA, but not Paclitaxel. These studies have recapitulated the effects of chemotherapy on various aspects of cognitive functioning in non-tumor rodent models, suggesting that in addition to the effects of cancer, chemotherapeutic agents themselves may be associated with neuropsychological deficits.

A variety of behavioral tests were used to assess cognitive functioning in rodent models. While there are many differences across studies, as seen in table

9, some tasks were more consistently associated with cognitive deficits than other tasks. A majority of studies using passive/inhibitory avoidance, novel location recognition, or Morris water maze tests reported chemotherapy-related impairments (Fardell, Vardy, Logge, & Johnston, 2010; Fardell, Vardy, Shah, &

Johnston, 2012; Konat, Kraszpulski, James, Zhang, & Abraham, 2008; Li et al.,

2008; Liedke et al., 2009; Lyons, ElBeltagy, Bennett, & Wigmore, 2012;

Madhyastha, Somayaji, Rao, Nalini, & Bairy, 2002; Mondie, Vandergrift, Wilson,

Gulinello, & Weber, 2010; Mustafa, Walker, Bennett, & Wigmore, 2008; Reiriz et al., 2006; Winocur, Binns, & Tannock, 2011; Winocur et al., 2012; Winocur,

Vardy, Binns, Kerr, & Tannock, 2006; M. Yang et al., 2010); (c.f. (Fremouw,

Fessler, Ferguson, & Burguete, 2012; Gandal, Ehrlichman, Rudnick, & Siegel,

2008; G. D. Lee et al., 2006; Long et al., 2011; Seigers et al., 2009), whereas most studies using fear conditioning did not (Fremouw, Fessler, Ferguson, &

Burguete, 2012; Gandal, Ehrlichman, Rudnick, & Siegel, 2008; Long et al.,

2011). The novel object recognition test resulted in mixed results (Fardell, Vardy,

Logge, & Johnston, 2010; Fardell, Vardy, Shah, & Johnston, 2012; Fremouw,

Fessler, Ferguson, & Burguete, 2012; Gandal, Ehrlichman, Rudnick, & Siegel,

2008; Y. Li, V. Vijayanathan, M. E. Gulinello, & P. D. Cole, 2010; Mondie,

Vandergrift, Wilson, Gulinello, & Weber, 2010; Seigers et al., 2009; M. Yang et al., 2010). As chemotherapy is associated with some domains of cognitive functioning more than others in breast cancer patients, the same is most likely true in rodent studies, as seen by the different response to chemotherapy across the various test. Furthermore, some tests may be more sensitive to chemotherapy-induced alterations than others. Future studies will be needed to further address this issue.

In keeping with trends in clinical research, most rodent studies of behavioral side effects of chemotherapy have focused on cognitive function; as seen in table 10, very few have assessed effects on affective behavior. Four of the eight studies listed in table 10 used the elevated plus maze (EPM) to assess anxiety-like behavior after administering doxorubicin. Three of the four found that chemotherapy did induce anxiety-like behavior, the fourth (Kujjo et al., 2011) only analyzed risk-assessment behavior, instead of the commonly reported time spent in open arms (Wall & Messier, 2001). Only one study assessed the anxiogenic effects of cyclophosphamide (Reiriz et al., 2006), reporting no drug effect.

However, this study used open field to assess anxiety-like behavior, but did not provide the data or include sufficient details regarding the parameters used to make an independent judgment. Two studies administering MTX reported conflicting results; one study reported increased freezing during fear conditioning

(Gandal, Ehrlichman, Rudnick, & Siegel, 2008) while the other found no difference using the light/dark box (Madhyastha, Somayaji, Rao, Nalini, & Bairy,

2002). Experimental designs were not ideal for these studies, however, as fear conditioning is a test used for learning and memory and mice in the second study were allowed 4 days of acclimation in the light/dark box although this test is generally used without acclimation (Hascoet & Bourin, 2009). Only two studies assessed depressive-like behavior (Kujjo et al., 2011; Mondie, Vandergrift,

Wilson, Gulinello, & Weber, 2010). Doxorubicin was shown to increase depressive-like behavior in the forced swim but not tail suspension test (Kujjo et

27 al., 2011), whereas thioTEPA did not increase depressive-like behavior in the forced swim or tail suspension tests (Mondie, Vandergrift, Wilson, Gulinello, &

Weber, 2010).

While very few studies have assessed the effects of chemotherapy on affective behavior in rodent models, the findings suggest that chemotherapy does induce anxiogenic, and possible depressive-like, behavior in rodents. However, only one study attempted to reverse the chemotherapy-induced effects on affective behavior, finding that alprazolam (a benzodiazepine) reversed the anxiety-like behavior associated with doxorubicin (Anwar, Pillai, Khanam, Akhtar,

& Vohora, 2011). Similarly numerous studies have assessed the mechanistic effects of chemotherapy in animal models without testing behavioral outcomes, as discussed in the next section. It is imperative that future studies address the mechanism and behavior together and attempt to reverse the cellular and behavioral changes induced by chemotherapy.

6. Mechanism

It has long been hypothesized that chemotherapy agents may have neuropsychological effects in non-CNS tumor cancer patients. Over the past decade findings from longitudinal studies in breast cancer patients and non- tumor animal models have provided support for these hypotheses, as previously discussed. However, not all patients receiving chemotherapy experience these side effects, with incidence rates ranging from 15%-80% (Ahles et al., 2010;

Koppelmans et al., 2012; Wefel, Saleeba, Buzdar, & Meyers, 2010).

Furthermore, these side effects are transient for some patients, while for others they can persist for more than a decade (Yamada, Denburg, Beglinger, &

Schultz, 2010). In recent years, a growing number of clinical and pre-clinical studies have attempted to assess the cause by which chemotherapy induces neurological deficits. Although the mechanism has not been elucidated, clinical studies utilizing imaging techniques and peripheral immune measures, in addition to pre-clinical studies (in-vivo and in-vitro), have identified specific effects and potential biological mechanisms by which chemotherapy is hypothesized to hamper neuropsychological functioning.

6.1 Immune-signaling in clinical studies

Recent literature suggests that people suffering with mild cognitive impairment have elevated levels of circulating pro-inflammatory cytokines

(Alvarez, Cacabelos, Sanpedro, Garcia-Fantini, & Aleixandre, 2007; Magaki,

Mueller, Dickson, & Kirsch, 2007) and that this pro-inflammatory bias may be associated with cognitive deficits in cancer patients (Ahles & Saykin, 2007;

Cleeland et al., 2003; Penson et al., 2000; Wang et al., 2010). Furthermore, some chemotherapy agents induce a peripheral pro-inflammatory response in an animal model (Tangpong et al., 2006), thereby providing support for the hypothesis that chemotherapy-induced elevation in inflammatory markers may play a role in neuropsychological deficits following chemotherapy treatment.

Indeed, the side effects of immunotherapy, treatment with pro-inflammatory cytokines, include cognitive deficits, depression, and fatigue (Trask, Esper, Riba,

& Redman, 2000). Although few clinical studies have assessed the immune response to chemotherapy treatment in breast cancer patients, findings overwhelmingly suggest that patients receiving chemotherapy do indeed express higher levels of certain peripheral pro-inflammatory markers than in BC patients who do not receive chemotherapy and healthy controls (Table 11). Increases in circulating interleukin-6 (IL-6) and tumor necrosis factor alpha (TNF-α), including its receptor soluble TNF receptor type 2(sTNFrII) are most commonly associated with chemotherapy (Janelsins et al., 2012; Pusztai et al., 2004; Torres et al.,

2013; Tsavaris, Kosmas, Vadiaka, Kanelopoulos, & Boulamatsis, 2002). In addition to assessing chemotherapy-related changes in cytokine expression, recent studies have also assessed functional outcomes related to these changes in cytokine levels (Bower et al., 2013; S. Kesler et al., 2013). Compared to healthy controls, breast cancer patients had higher levels of TNF-α and IL-6, memory impairment, and reduced hippocampal volume in the left hemisphere (S.

Kesler et al., 2013); a chemotherapy-related reduction in size of the right hippocampus approached significance (p=.07). Furthermore, total memory functioning (Hopkins Verbal Learning Test (HVLT) total) and hippocampal volume was associated with TNF-α and IL-6 in the chemotherapy group, but not the healthy controls. Although this study did not include a non-chemotherapy breast cancer control group, they did report that TNF-α concentrations decreased

30 as time since the final chemotherapy treatment increased (S. Kesler et al., 2013).

Furthermore, there was no relationship between stage of breast cancer and soluble TNF-α concentrations, suggesting that these changes are related to chemotherapy treatment and not the cancer (S. Kesler et al., 2013). Together, these data suggest that chemotherapy-induced cognitive deficits and changes in brain structure may be related to chemotherapy-related changes in TNF-α and

IL-6.

A longitudinal study using a BC control group that did not receive chemotherapy reported higher levels of soluble TNF-α receptor type 2 (sTNF-RII) in the chemotherapy group shortly after, and 6 months following, treatment with no difference 1 year later (Ganz et al., 2013). PET imaging also revealed that sTNF-RII is associated with diminished metabolism in the inferior frontal cortex and memory complaints in the chemotherapy group, but not the control group.

The longitudinal decline in sTNF-RII was also correlated with fewer memory complaints over twelve months. Although there was no relationship between cytokines and non-self-reported cognitive functioning, the neuropsychological outcomes have not been assessed beyond baseline as the full study has not yet been completed. A third study, also incorporating non-chemotherapy breast cancer controls reported that chemotherapy, but not radiotherapy, was associated with higher levels of NF-kB DNA binding in peripheral blood mononuclear cells, IL-6, and sTNF-R2, as well as depression and fatigue (Torres

31 et al., 2013). Chemotherapy and NF-kB DNA binding were also significant predictors of depression.

While these findings support an inflammatory based mechanism by which chemotherapy alters neuropsychological functioning, more research is needed to assess a long-term presence and influences of these cytokines and to determine if a causal relationship exists between chemotherapy-induced inflammation and changes in brain structure and function. Only two longitudinal studies have assessed cytokine concentrations out to one year after treatment, both finding a decrease in sTNFrII across time (Ganz et al., 2013; Pomykala et al., 2013), while

C-reactive protein (CRP) remained higher in chemotherapy treated patients in one study (Pomykala et al., 2013), the other finding no treatment differences in

CRP at any time point (Ganz et al., 2013). One study also suggested that findings may be drug specific; concentrations of IL-6 were higher in breast cancer patients treated with AC/CAF compared to CMF (Janelsins et al., 2012). This may be due to the anti-inflammatory properties of methotrexate (Johnston,

Gudjonsson, Sigmundsdottir, Ludviksson, & Valdimarsson, 2005). However, removing the participants who had received methotrexate did not alter the results in a similar study (S. Kesler et al., 2013). These limited findings suggest that chemotherapy-induced inflammation is correlated with neuropsychological effects seen in breast cancer patients treated with chemotherapy agents.

6.2 Pre-clinical studies

Findings from numerous clinical studies suggest that various chemotherapy agents are associated with neuropsychological deficits.

Furthermore, imaging studies have found that patients treated with chemotherapy have structural and functional changes in their brains compared to healthy controls as well as non-chemotherapy treated breast cancer patients.

Pre-clinical studies corroborate and extend these findings, reporting chemotherapy-induced neuropsychological changes in non-tumor bearing rodents. However, even with this vast amount of literature, the mechanism by which this occurs remains unknown. Consistent results in pre-clinical and clinical research suggest that neurobiological research in animal models may be useful in delineating the mechanisms through which chemotherapy alters behavior. In the past decade, a growing number of pre-clinical studies have proposed potential mechanisms associated with chemotherapeutic neurotoxicity within the

CNS, but no interventional studies have attempted to establish causation.

As described above, a wide range of chemotherapeutic agents have been linked to cytotoxic damage within the CNS (see Table 13), including doxorubicin, cyclophosphamide, methotrexate, 5-fluorouracil, paclitaxel, thioTEPA, carmustine, cytarabine, cisplatin, ifosfamide, temozolomide, vinblastin, cytosine arabinoside, and vincristine (Dietrich, Han, Yang, Mayer-Proschel, & Noble,

2006; Eijkenboom & Van Der Staay, 1999; Han et al., 2008; Janelsins et al.,

2010; Koros & Kitraki, 2009; Rzeski et al., 2004; Tangpong et al., 2006; M. Yang

33 et al., 2011). Furthermore, these drugs alter a variety of biological components within the brain including neurogenesis, cell death, oxidative stress, inflammation, neurotransmitters and monoamines (Aluise et al., 2011; Aluise et al., 2010; Briones & Woods, 2011; Han et al., 2008; Madhyastha, Somayaji, Rao,

Nalini, & Bairy, 2002; Tangpong et al., 2006).

Doxorubicin has been shown to increase CNS inflammation, cell death, apoptosis, oxidative stress and impair mitochondrial functioning (Joshi et al.,

2007; Kang, Chess-Williams, Anoopkumar-Dukie, & McDermott, 2013; Oz &

Ilhan, 2006; Tangpong et al., 2011). Rodents treated with doxorubicin had higher concentrations of pro-inflammatory cytokine TNF-α, increased levels of inducible nitric oxide synthase (iNOS) mRNA and more oxidative stress as indicated by increased protein and lipid oxidation and decreased glutathione (GSH; Aluise et al., 2011; Joshi et al., 2007; Tangpong et al., 2007; Tangpong et al., 2006).

Administration of an antioxidant, mercaptoethane sulfonate (MENSA), prevented the changes in TNF-α and oxidative stress (Aluise et al., 2011). The increase in

TNF-α may be associated with mitochondrial dysfunction via oxidative stress as doxorubicin increased TNF-α in wild-type and iNOS knock-out mice, but only affected mitochondrial respiration in the WT mice (Tangpong et al., 2007).

Furthermore, inhibiting TNF-α ameliorated the increase in iNOS mRNA and mitochondrial dysfunction (Tangpong et al., 2006). Future studies using doxorubicin need to assess behavioral outcomes in addition to oxidative stress and inflammation to validate the biological relevance of these findings.

Cyclophosphamide is associated with alterations in neurogenesis, oxidative stress, cell death, and histone acetylation (Briones & Woods, 2011;

Manda & Bhatia, 2003; Rzeski et al., 2004; M. Yang et al., 2010). Rodents treated with cyclophosphamide showed a decrease in neurogenesis which was also associated with cognitive impairment in the novel object recognition, novel location recognition, Morris water maze, and fear conditioning tests (Briones &

Woods, 2011; Christie et al., 2012; Janelsins et al., 2010; M. Yang et al., 2010).

Similar to doxorubicin, cyclophosphamide treatment resulted in decreased levels of glutathione and glutathione peroxidase and increased levels of malondialdehyde, aspartate aminotransferase, and alanini aminotransferase, indicative of oxidative stress (Bhatia, Manda, Patni, & Sharma, 2006; Manda &

Bhatia, 2003; Oboh & Ogunruku, 2010). Although these studies assessing cyclophosphamide associated oxidative stress did not attempt to test neuropsychological deficits, a study assessing oxidative stress and neuropsychological outcome reported that treatment with methotrexate increased oxidative stress, depressive-like behavior, and cognitive dysfunction. Thus, chemotherapy-associated increases in oxidative stress may have an effect on neuropsychological functioning. Cyclophosphamide-treated rats also had histone modifications in the hippocampus and prefrontal cortex, but not striatum, relative to control rats (Briones & Woods, 2011). Treatment resulted in increased acetylation of histone H3 and decreased histone deacetylation activity,

35 suggesting that cyclophosphamide may influence epigenetic modifications within the brain.

As seen in Table 13, oxidative stress is commonly studied in and associated with doxorubicin and cyclophosphamide and could be one mechanism contributing to chemotherapy-related cognitive impairment (Ahles &

Saykin, 2007). Oxidative stress is a common source of DNA damage that occurs due to secondary effects of metabolism and exogenous toxins (Fishel, Vasko, &

Kelley, 2007). Doxorubicin and cyclophosphamide are effective cancer treatments, in part because they disrupt the DNA. These effects are not cell specific, however, and cause similar damage to other cells. As chemotherapy is associated with cognitive deficits, similarly, DNA damage is higher in elderly patients diagnosed with mild cognitive impairment compared to healthy controls(Keller et al., 2005). These findings suggest that oxidative stress is a logical mechanism for chemotherapy induced neuropsychological changes and warrants further research.

Pre-clinical studies using methotrexate have found this drug to be associated with a wide range of alterations in the brain including neurogenesis, gliogenesis, oxidative stress, apoptosis, blood vessel density, CSF composition, electrophysiology, and neurotransmitters (Gandal, Ehrlichman, Rudnick, &

Siegel, 2008; Madhyastha, Somayaji, Rao, Nalini, & Bairy, 2002; Rajamani,

Muthuvel, Senthilvelan, & Sheeladevi, 2006; Seigers et al., 2008; Silverstein &

Johnston, 1986; Vijayanathan, Gulinello, Ali, & Cole, 2011; M. Yang et al., 2012;

M. Yang et al., 2011). Methotrexate also is associated with suppressed neurogenesis and gliogenesis of the hippocampus, specifically the subgranular zone, and related deficits in learning and memory (Lyons et al., 2011; Seigers et al., 2010; Seigers et al., 2008; Vijayanathan, Gulinello, Ali, & Cole, 2011; M.

Yang et al., 2012; M. Yang et al., 2011). Interestingly, in vitro studies suggest that susceptibility to methotrexate induced cytotoxicity or apoptosis depends on the maturity of the cells; cytotoxicity of immature hippocampal cells increases in a dose-dependent manner whereas mature cells do not change. Furthermore, immature cells also have increased expression of active caspase-3 and cleaved poly (ADP-ribose) polymerase (PARP), markers of apoptosis (M. Yang et al.,

2011). This same study reported that pretreatment with a caspase-3 inhibitor blocked methotrexate-induced hippocampal cell death, suggesting that methotrexate may induce apoptosis of immature hippocampal cells.

Methotrexate is also associated with changes in cerebral spinal fluid

(CSF) components. Folate is a B vitamin found in the blood and CSF that is essential for brain development and optimal cognitive functioning. It is also involved in neurotransmitter synthesis and metabolism of homocysteine, a neurotoxic amino acid (Y. Li, V. Vijayanathan, M. Gulinello, & P. D. Cole, 2010).

Low circulating folate is associated with mild cognitive impairment and depression (Bell et al., 1992; Riggs, Spiro, Tucker, & Rush, 1996). In pre-clinical studies MTX administration, both acute and chronic, was associated with reduced folate in the CSF and serum (Y. Li, V. Vijayanathan, M. Gulinello, & P.

D. Cole, 2010; Y. Li, V. Vijayanathan, M. E. Gulinello, & P. D. Cole, 2010).

Methotrexate is also associated with reduction of monoamines including dopamine, norepinephrine, and serotonin (Madhyastha, Somayaji, Rao, Nalini, &

Bairy, 2002; Silverstein & Johnston, 1986). These are important findings as these monoamines play critical roles in depression and anxiety (Dunlop & Nemeroff,

2007). Methotrexate is involved in a number of different possible mechanisms underlying neuropsychological deficits. Future research needs to address whether these different components are related to or independent of each other.

Although less extensively studied, 5-FU, thioTEPA, cytosine arabinoside, carmustine, cisplatin, and cytarabine are also associated with cytotoxicity in the brain (Dietrich, Han, Yang, Mayer-Proschel, & Noble, 2006; Han et al., 2008;

Mondie, Vandergrift, Wilson, Gulinello, & Weber, 2010). 5-FU is associated with reduced hippocampal cell proliferation and impaired cognitive functioning (NLR;

ElBeltagy et al., 2010; Lyons, ElBeltagy, Bennett, & Wigmore, 2012; Mustafa,

Walker, Bennett, & Wigmore, 2008). Fluoxetine, an anti-depressant, improved the treatment-associated behavioral and proliferation deficits (ElBeltagy et al.,

2010; Lyons, ElBeltagy, Bennett, & Wigmore, 2012). Findings from an in vitro study suggest that CNS cells are more susceptible to chemotherapy agents than cancer cells (Han et al., 2008); chemotherapeutic agents are known to inhibit proliferation, however, treatment of 5-FU killed a majority of CNS cells in vitro without harming tumor cells. Both mature oligodendrocytes and their precursors, essential for myelinating axons, were destroyed. In vivo, chemotherapy caused a

2.5 fold increase in apoptosis in the subventricular zone and a four-fold increase in apoptosis in the dentate gyrus of the hippocampus one day after treatment, with increased cell death continuing for two weeks after treatment with 5-FU (Han et al., 2008). Furthermore, chemotherapy treatment caused delayed abnormalities in transcriptional regulation and maintenance of myelin integrity, causing myelin loss months after treatment ended (Han et al., 2008).

Carmustine, cisplatin, and cytarabine had similar effects, being more toxic for

CNS progenitor cells and non-dividing oligodendrocytes than cancer cells, with cell death increasing for weeks after treatment cessation (Dietrich, Han, Yang,

Mayer-Proschel, & Noble, 2006). These findings offer a possible explanation for the long-term neuropsychological effects of chemotherapeutic agents.

Pre-clinical studies have been able to reproduce the neuropsychological effects of chemotherapy seen in cancer patients and assess the effects of chemotherapeutic agents individually, whereas a majority of clinical studies involve multiple chemotherapeutic agents. Although the mechanism has yet to be elucidated, findings from pre-clinical studies clearly show possible components that may be involved including neurogenesis/gliogenesis, apoptosis, inflammation, oxidative stress, CSF composition, and monoamines. Potential interventions have also been suggested including administration of anti-oxidants

(Aluise et al., 2011; Oboh, Akomolafe, Adefegha, & Adetuyi, 2012; Tangpong et al., 2007), stimulating neurogenisis (Janelsins et al., 2010), and anti-depressant

Fluoxetine (ElBeltagy et al., 2010). However, these interventions have not yet

39 been used in a tumor-bearing model, which may result in a different outcome. It is also difficult to determine which mechanistic components have a direct versus indirect effect on neuropsychological ability and the extent to which they are inter-related. Furthermore, pre-clinical research has yet to address the observation that impairment only occurs in a subset of patients. Future studies will need to address this issue, looking at possible genetic variances including drug resistance pumps, DNA repair mechanisms and cytokine/hormone production and receptors (Ahles & Saykin, 2007).

7. Pre-clinical Roadblocks

Many cancer patients treated with chemotherapeutic agents experience neuropsychological deficits, with reported incidences of 15% to 80% (Ahles et al.,

2010; Bjorneklett et al., 2012; Koppelmans et al., 2012; So et al., 2010; Wefel,

Saleeba, Buzdar, & Meyers, 2010). Although numerous clinical studies have assessed the relationship between chemotherapeutic agents and neuropsychological deficits, as previously discussed, a mechanism has not been elucidated. In the past decade, a growing number of rodent models have emerged, greatly increasing our understanding of possible key components involved in this relationship. Most importantly, animal models were able to recapitulate the chemotherapy related neuropsychological deficits reported in clinical studies. However, basic research in this area began only recently and is still in the early stages. Future studies will need to be more mechanistic and

40 focus more on the cellular and molecular components and pathways involved.

Multiple challenges or roadblocks in animal models emerge as there are experimental design differences across studies that may result in various findings, in turn hindering translational outcomes for cancer patients. These challenges are discussed below.

7.1 Rodent models

When designing a translational study, it is important to determine the degree of translational relevance related to a given model. Rodent models vary greatly in areas that may hinder their translational value to cancer patients, such as age, sex, and hormone concentrations. A majority of breast cancer patients are 50 years old or older, only five percent being younger than 40 ("Breast

Cancer Facts & Figures 2011-2012"). However, most pre-clinical research uses young adult rodents (see Table 13). Clinical studies suggest that age is associated with decreased cognitive ability (Verhaeghen & Salthouse, 1997).

Furthermore, pre-clinical studies report age-related changes in the brain that are associated with increased inflammation and depressive-like behavior (Corona,

Fenn, & Godbout). In order to better mimic the clinical breast cancer population, thus potentially having more translational relevance, future studies should examine whether the central and behavioral effects of chemotherapy differ in young adult and aged adult mice.

Choosing the appropriate sex is important in pre-clinical research as sex differences exist in both immune function and cognitive ability (Jonasson, 2005;

Klein & Nelson, 1998; Nava-Castro, Hernandez-Bello, Muniz-Hernandez,

Camacho-Arroyo, & Morales-Montor, 2012). Males generally have reduced immune responses compared to females, likely due to the testosterone suppressing the immune system (Nava-Castro, Hernandez-Bello, Muniz-

Hernandez, Camacho-Arroyo, & Morales-Montor, 2012; Olsen & Kovacs, 1996).

Indeed, male mice had a higher immune response after infection if their gonads had been removed compared to those with intact gonads (Kittas & Henry, 1980).

Similarly, male rats are more susceptible to Adriamycin-induced nephropathy than females and castrated males were less susceptible than sham-operated controls (V. W. Lee & Harris, 2011). Sex differences also exist in cognitive functioning in rodents. A meta-analysis reported that male rats performed better than females in the water maze and radial arm maze, tasks assessing spatial, working, and reference memory (Jonasson, 2005). In contrast, female mice performed better in the water maze than male mice, while male mice still performed better than female mice in the radial maze. These findings suggest that results may vary across studies based on the species and sex of animal used as immune function and cognitive ability vary across species and sex.

Furthermore, when choosing which sex to use, it is important to determine the desired translational goal of the study. As breast cancer is the second most common form of cancer, is female dominant, and has a high survival rate, most

42 clinical research has used breast cancer patients when assessing the side- effects of chemotherapy (American Cancer Society. Cancer Facts & Figures

2013. Atlant: American Cancer Society; 2013, 2013). Thus, it would be most beneficial to use female animal models for translational research In future studies. It is essential that these aspects are accounted for and understood before beginning an experiment, as the species and sex of rodent used may be just as important as choosing which chemotherapeutic agent or behavioral test to use and will have varying effects on translational outcomes.

The effect of hormones on cognitive functioning is also important in animal models as clinical studies suggest that changes in hormone levels and menopausal status may alter cognitive ability (Berent-Spillson et al.; Jenkins,

Shilling, Fallowfield, Howell, & Hutton, 2004). Animal models assessing the neuropsychological effects of chemotherapy vary in this regard as some use ovariectomized females to better mimic post-menopausal breast cancer patients

(Macleod et al., 2007), while others leave the ovaries intact (Winocur, Vardy,

Binns, Kerr, & Tannock, 2006). To our knowledge no pre-clinical chemotherapy studies have assessed neuropsychological functioning comparing ovariectomized and non-ovariectomized rodents. This would be beneficial in determining if the effects of specific chemotherapy agents vary according to hormones or if hormone levels, one way or the other, exacerbates the effects of chemotherapy on neuropsychological outcomes. As previously mentioned, assessing the desired translational relevance is important when designing an

43 animal model. A majority of breast cancer patients are peri- or post-menopausal before beginning chemotherapy treatment (Ahles et al., 2010; Collins,

MacKenzie, Tasca, Scherling, & Smith, 2012; Del Mastro et al., 2006; Jenkins et al., 2006; Shilling, Jenkins, Morris, Deutsch, & Bloomfield, 2005), less than 25% being pre-menopausal (Bines, Oleske, & Cobleigh, 1996; Hankey, Miller, Curtis,

& Kosary, 1994). Furthermore, chemotherapeutic agents frequently induce early menopause in premenopausal breast cancer patients (Bines, Oleske, &

Cobleigh, 1996; Del Mastro et al., 2006). As the main goal of related pre-clinical research is to improve the quality of life in chemotherapy treated patients by suppressing treatment-related neuropsychological deficits, using ovariectomized females would be most beneficial in translational outcomes.

A number of different strains of rats and mice have been used to study the mechanism of chemotherapy-related neuropsychological deficits including BalbC,

C57bl/6, Swiss albino, ICR, and B6C3 mice, and Wistar, Lister-hooded,

Sprague-Dawley, athymic nude, Buffalo, and Long Evans rats. Although these species and strains are similar in many instances and have all shown deficits related to chemotherapeutic agents, there are differences among them that may result in different outcomes in future chemotherapy studies. For example, differences in neurogenesis are seen between rats and mice as newly forming hippocampal neurons are more numerous, faster-maturing, and more likely to be recruited to learning circuits in rats than in mice; contributions of these new neurons to fear memory are also greater in rats than mice (Snyder et al., 2009).

Whether, chemotherapy has similar effects on neurogenesis in rats and mice remains to be determined.

Another important factor could be the immune system, as it is hypothesized to play a role in chemotherapeutic side effects (Kang, Chess-

Williams, Anoopkumar-Dukie, & McDermott, 2013; Tangpong et al., 2007) and variations in immune response exist between strains (Griffin & Whitacre, 1991;

Palumbo et al., 2010; I. V. Yang et al., 2009) and species of rodents (Yang,

2009). At basal levels Lewis rats have higher mononuclear cell counts and CD4 lymphocytes than Fischer (F344) rats (Griffin & Whitacre, 1991). Furthermore stress induces different immune responses across strains, generating a Th2 response in BALB/c and Th1 response in C57BL/6 mice; these responses were marked by a difference in cytokine expression (Palumbo et al., 2010). For example, BALB/c mice have increased IL-4 and IL-10 and decreased INF-γ concentrations compared to C57BL/6 mice (Palumbo et al., 2010). Similarly, treating sixteen different mouse strains with LPS for 6 hours resulted in a wide range of cytokine concentrations depending upon the strain (I. V. Yang et al.,

2009). Differences were also apparent between rats and mice with cells from the small intestine of C57BL/6 mice having higher concentrations of TNF-α, IL-4, and

GM-CSF, whereas Sprague-Dawley rats had higher concentrations of IL-1b (I. V.

Yang et al., 2009). These differences in immune regulation could potentially result in chemotherapeutic agents having differing effects across species and strains. Indeed, in assessing adriamycin-induced nephropathy, C57BL/6 mice

45 were much more resistant to renal injury than BALB/c mice, needing a higher dose to provoke injury (V. W. Lee & Harris, 2011).

As seen in Table 13 most pre-clinical research has assessed biological and/or behavioral outcome in a non-tumor model using a single chemotherapeutic agent. Using non-tumor models has been beneficial in assessing the effect of chemotherapy on neuropsychological functioning independent of cancer and will continue to be paramount in elucidating the mechanism. However, as therapeutic measurements arise it will be essential to assess these treatments in a tumor-bearing model as it is well documented that cancer alters immune functioning and behavior (Wefel et al., 2004; M. Yang et al., 2010). Furthermore, tumor models in which the tumor is removed before administering chemotherapy will also be essential, as this models clinical adjuvant treatment regimens. Similarly, assessing the effects of individual chemotherapeutic agents has been and will be essential to understanding the mechanistic relationship between these drugs and related neuropsychological deficits. However, as cancer patients are frequently treated with multiple drugs, some pre-clinical experiments will also need to include these combined cocktails to assess the possible effect of drug interactions on cognition and behavior.

7.2 Drug Administration

Chemotherapeutic agents in breast cancer patients are commonly administered intravenously, rapidly circulating throughout the body. Each

46 treatment generally consists of multiple cycles administered every few weeks. In clinical studies, dosage and number of treatment cycles is positively correlated with neuropsychological deficits (Hodgson, Hutchinson, Wilson, & Nettelbeck,

2013). As it is difficult to mimic the slower drip-line administration of chemotherapeutic agents seen in cancer patients, rodent models generally administer a bolus injection intraperitoneally (i.p.; Han et al., 2008; Rzeski et al.,

2004), or intravenously (i.v.,; Helal et al., 2009; Seigers et al., 2010); i.p. administration is more commonly used than i.v. administration. Intraperitoneal injections mimic the slower administration of clinical patients as the drugs slowly diffuse across the abdominal wall and into the blood stream. However, i.p. administration is also more toxic at the site of injection and results in a smaller concentration of drug reaching the blood stream (Aletti et al., 2010). In a clinical study of women with advanced ovarian cancer, women receiving i.p. chemotherapy had more abdominal pain, nausea, vomiting, and other side effects than the women receiving treatment intravenously (Armstrong et al.,

2006; Wenzel, Huang, Armstrong, Walker, & Cella, 2007).

Intravenous administration in rodents mimics clinical route of administration. However, bolus injections are given instead of a drip line, resulting in a much larger immediate concentration of chemotherapeutic agents in the blood stream. Some studies have addressed this by administering multiple injections of smaller drug concentrations over a short period of time (Rajamani,

Muthuvel, Senthilvelan, & Sheeladevi, 2006). It is not known if these alterations

47 in drug administration result in different chemotherapy associated changes in neuropsychological functioning. A few studies have also used intracerebroventricular or intrathecal injections in order to determine the direct effect of chemotherapeutic drugs in the brain, bypassing drug metabolism in the liver and blood brain barrier protection (Lau et al., 2009; Vijayanathan, Gulinello,

Ali, & Cole, 2011). However, this approach is not very relevant to assessing chemotherapeutic-related effects on cognition and affect as it is very difficult for most chemotherapy agents to cross the blood brain barrier, this being a major obstacle when trying to treat CNS tumors (Gerstner & Fine, 2007; Sawyer,

Piepmeier, & Saltzman, 2006).

Another potential roadblock in pre-clinical research is the vast variation in dosage, frequency and duration of chemotherapeutic treatment. As seen in Table

13, the variation in dosage is small for some drugs such as doxorubicin and cyclophosphamide, while it is much larger for others such as Methotrexate, ranging from 0.1 mg/kg to 250 mg/kg for a single treatment. Additionally, rodent studies vary greatly in the number and frequency of treatments. The studies shown in Table 13 range from one to ten injections with frequency ranging from once each day to once each week and duration from first treatment to analysis ranging from 3 hours to eight weeks (Y. Li, V. Vijayanathan, M. E. Gulinello, & P.

D. Cole, 2010; Tangpong et al., 2006). These differences may result in varying results, as both dose and number of chemotherapy treatments are associated with worse neuropsychological outcome (Hodgson, Hutchinson, Wilson, &

Nettelbeck, 2013; Schagen, Muller, Boogerd, Mellenbergh, & van Dam, 2006). In determining a proper dose, it is recommended to calculate the human-equivalent- dose (HED) which is based on surface area instead of height and eight (Reagan-

Shaw, Nihal, & Ahmad, 2008). Using similar dosages and frequency and duration of treatments will be beneficial in comparing across future studies as the cellular and molecular mechanisms being tested become more specific and complex.

In the past decade, rodent models have increasingly been used to elucidate the mechanism by which chemotherapeutic agents are associated with neuropsychological alterations. Animal models are able to replicate the chemotherapy-associated deficits seen in clinical studies. However, as research in this area becomes more mechanistic, focusing on cellular and molecular pathways, it will be essential to understand how the differences in species, sex, tumor presence and chemotherapeutic agents might affect cellular, molecular, and immunological pathways differently and to design the studies with these in mind.

As discussed in a review on chemotherapy and animal models (Seigers,

Schagen, Van Tellingen, & Dietrich, 2013), future research would benefit greatly from having a shared database in which methods and study designs for all related studies were entered in and readily accessed by everyone. It would also be beneficial for researchers to be able to access negative findings in addition to positive outcomes. For example, researchers might have tried multiple dosages

49 and frequencies with a specific drug in a specific species before finding one that worked. Having access to this information would facilitate future research.

An important future direction for the preclinical work will be to understand individual variability in susceptibility to the cognitive and affective side-effects of chemotherapy. For some patients, these deficits never emerge, while for others they emerge and are resolved within a year, while for others still, they will persist for decades. A number of psycho-social factors have been postulated as contributing to individual differences in susceptibility to these side effects of treatment, including pre-treatment cognitive ability, socioeconomic status, education, age, and social support (Ahles et al., 2002; Avis et al., 2012; Caplette-

Gingras & Savard, 2008; Cimprich, So, Ronis, & Trask, 2005; S. R. Kesler et al.,

2013; Kissane et al., 2004), but the relative importance of each of these variables in cognitive and affective deficits remains to be determined. In addition, it has recently been suggested that genetic variations may also be involved; indeed, single nucleotide polymorphisms in promoter regions of cytokine DNA are associated with increased neuropsychological deficits (Bower et al., 2013).

8. Conclusions

Chemotherapy treatment is frequently administered to breast cancer patients to prevent tumor recurrence and has been beneficial in increasing the overall 10-year survival rate, now at 84.5% (Howlader N, 2011). However,

50 chemotherapeutic agents are also associated with neuropsychological impairment and can have a negative impact on quality of life. Although there are over 1,000 publications examining these side effects, the mechanism by which chemotherapeutic agents are associated with neuropsychological deficits has not been elucidated. This is likely due to the wide variations in study design, cognitive tests, chemotherapy drugs, treatment duration and method of data analysis. However, imaging studies have left little doubt that patients treated with chemotherapy have reduced volume of both white and gray matter, and altered brain activity during cognitive tasks. Furthermore, imaging studies have been more consistent in documenting cognitive deficits than those not utilizing imaging. Future research needs to assess the sensitivity of the tests currently used for cognitive assessment in chemotherapy treated patients.

In the past decade, a growing number of pre-clinical studies using non-tumor bearing rodent models have corroborated clinical findings, reporting that rodents treated with chemotherapy have cognitive deficits and show increased depressive-like and anxiety-like behavior. While it is difficult to assess the effect of each individual drug in clinical research, rodent models have found that many different chemotherapy agents are independently associated with neuropsychological deficits. Furthermore, these studies have proposed potential mechanisms associated with chemotherapeutic neurotoxicity within the CNS involving neurogenesis, cell death, oxidative stress, inflammation,

51 neurotransmitters and monoamines. However, no interventional studies have attempted to establish causation.

Until the past decade, a majority of clinical studies were descriptive, assessing correlative associations between chemotherapy and cognitive deficits.

Accordingly, in 2011 the International Cognition and Cancer Task Force reiterated the importance of appropriate study design and recommended longitudinal studies, focusing on causation. Similarly, still in its infancy, pre- clinical studies have determined a correlation between chemotherapy and neuropsychological deficits, but have not focused on designing causational studies with an emphasis on translational outcome. Now that a number of possible mechanisms have been suggested using single chemotherapeutic agents, it will be important for future study designs to include the age, sex, hormone status, and drug combination, administration and dosage that will best mimic the clinical setting, with a focus on translational outcomes. Furthermore, a shared database containing detailed information about study design, methods, and results, both positive and negative, would be extremely valuable in facilitating progress in this field. Following these guidelines will assist in translating pre-clinical findings to clinical studies by reducing chemotherapy- associated neuropsychological deficits and increasing quality of life of cancer patients, which will likely reduce the recurrence rate and mortality rate of breast cancer.

Table 1. Self Report of Cognitive Deficits

Instrument Design n per group Controls Chemo Conclusion Notes/Limitations Ref

BCFQ Longitudinal 43-50 healthy FEC, CMF, Higher self report deficits for Preliminary data from ongoing Schilling baseline, 6 mo AC chemotherapy patients than controls, but study with later time point 2005 not correlated with objective scores MASRQ Longitudinal 45-60 BC-no ACP,AC, The increase in cognitive symptoms was 80% of chemo and 66% of BC-no Ahles baseline, 1 mo, 6 chemo, CAF,FEC, greater for chemotherapy patients than chemo groups on ET. Sample: 2010 mo, 18 mo healthy CMF patients not treated with chemotherapy primarily well-educated and healthy controls Caucasian women. PAOF Longitudinal 30-31 BC-no ACT, CMF, Self report of memory problems did not Older women (~61 yo).Women Tager baseline, after chemo, AC differ by treatment; however, they were who withdrew mid-study were 2010 chemo, 6 mo later associated with depression or anxiety more cognitively impaired than scores at various time points. the group mean on several objective measures. ET use varied

53 across time points and by group.

AFI Longitudinal 30 none AC Women report a significant reduction in None of women taking ET at time Jansen baseline, ~ 1-4 perceived cognitive performance from of testing. 2008 weeks after final baseline to post-treatment testing. chemo infusion FSCL Longitudinal 27 none AC/CAF, Proportion of patients reporting Among patients receiving Janelsins (cognitive cycle 2, cycle 4 CMF cognitive deficits higher for those AC/CAF changes in IL-6 were 2012 subset) receiving AC/CAF versus CMF during positively correlated with all 5 chemotherapy or immediately after cognitive measures, although not conclusion. approaching significance in any measure except difficulty thinking (p=0.059). Semi- Cross-sectional 19-34 BC-no CMF +/- Women treated with chemotherapy The cognitive score was highly Schagen Structured ~2 yr after start of chemo ET (CMF) had significantly lower scores correlated with the emotional 1999 Interview treatment relative to patients who did not receive functioning scale EORTC-QLQ. chemotherapy after surgery Dutch cohort. Continued

Table 1 continued

Instrument Design n per group Controls Chemo Conclusion Notes/Limitations Ref

SMSRQ Cross-sectional 22-35 BC-no 20 diff. The chemotherapy group reported No significant correlation Ahles ~10 years after chemo, regimens, greater deficits in working memory than between the SMSRQ subscales 2002 treatment lymphoma most the no chemotherapy group. There was a and domain scores. containing trend toward impaired new learning AC (p=0.07) but no significant effect on remote retrieval. FEDA Cross-sectional 23-29 BC-no EC, CMF, No differences in self-perceived Two chemotherapy groups more Mehnert ~5 years after chemo CTM cognitive deficit scores among standard likely to have had mastectomy 2007 treatment dose chemotherapy, high dose and higher tumor grade than non- chemotherapy, and non-chemotherapy chemotherapy controls. High-risk BC controls. Self perceived cognitive German cohort using German function does not correlate with versions of instruments (some objective assessments of function. modified). Abbreviations: AFI: Attentional Function Index; ET: endocrine therapy; BCFQ: Broadbent Cognitive Failures Questionnaire; MASRQ: Multiple Ability Self-Report

5 Questionnaire; EORTC-QLQ: European Organization for Research and Treatment of Cancer –Quality of Life Questionnaire. PAOF: Patient’s Assessment of Own

Functioning; SMSRQ: Squire Memory Self-Rating Questionnaire; FEDA: Questionnaire of Self-Perceived Deficits in Attention (German). FSCL: Fatigue Symptom Check List; AC: adriamyacin, cyclophosphamide; CAF: adriamycin, cyclophosphamide, 5-fluorouracil; CMF: cyclophosphamide, methotrexate, fluorouracil; ACT: adriamyacin, cyclophosphamide, and a taxane; FEC: fluorouracil, epirubicin, cyclophosphamide; DC: docetaxil, cyclophosphamide; taxC: a taxane and carboplatin; GC: gemcitabine, cisplatin; AP: adriamycin, paclitaxel; CFP: cyclophosphamide, 5-fluorouracil, paclitaxel; ET: endocrine therapy; EC: epirubicin, cyclophosphamide; CTM: cyclophosphamide, thiotepa, mitoxantrone; ACP: adriamyacin, cyclophosphamide, paclitaxel.

Table 2. Composite Scores for Global Cognitive Function

Instruments Design n per Controls Chemo Conclusion Notes/Limitations Ref group RBANS Longitudinal 30 none AC From baseline to post-chemo, None of women taking ET at time of Jansen baseline, ~ 1-4 reduction in total cognitive score testing. 2008 weeks after final and visuospatial skill; no diff. in chemo infusion immediate memory, language, attention, or delayed memory. CNS-VS, WAIS- Longitudinal 60 healthy FEC-T, Cognitive function is impaired at Progressive worsening of cognitive Collins III, PASAT, CCC, baseline and FEC, CT, several time points after function during chemotherapy 2013 COWA, HLVT-R, before each AC-T, AC chemotherapy. Working memory treatment rel. to controls. One year BVMT-R infusion and processing speed appear to be results not yet available. Canadian most sensitive to chemotherapy cohort. effects. PASAT; TMT A Longitudinal 40-53 BC-no FEC Higher rate of decline from All participants post-menopausal. Collins and B, WCST; baseline, 1 mo chemo (~50%), baseline to 1 month in the chemo Higher tumor grade in chemo group. 2009 BNT; COWA; GP; after chemo, 1 CEF, CAF, group (34% v 13%) ; chemo Canadian cohort. Women who DSC; WAIS-III; year after chemo AC, ACT, impaired working and visual withdrew mid-study were below the

5 CVLT II; WMS- ECT memory at 2nd time point; exec. mean on several cognitive measures III; RVLT; CCC function was improved one year at baseline; at baseline, chemo later in chemo group. Trend groups scored higher on several (p=0.06) for lower overall measures. summary in chemo patients receiving ET; sig. deficits were apparent in verbal memory and processing speed. Composite Score Longitudinal 19-34 BC-no CTC (high Greater cog. decline across time in High-risk Dutch cohort. Schagen from 10 tests baseline, 6 mo chemo, dose), FEC CTC patients 2006 (details absent) after chemo healthy controls WAIS-III, WRAT, Cross-sectional 22-35 BC-no 20 diff. Chemo impaired neuropsych. Correlation between the # of Ahles BNT, COWA, ~10 years after chemo, regimens performance; verbal memory and chemotherapy cycles and domain 2002 CVLT, WMS-R; treatment lymphoma psychomotor functions were most scores. No corr. between cognitive CPT affected. function and time since treatment. No differences in dep. or anx. among groups. 55 Continued

Table 2 continued

Instruments Design n per Controls Chemo Conclusion Notes/Limitations Ref group TMT, TAP, WMS- Cross-sectional 23-29 BC-no CMF, CMT Global cognitive impairment in 8% Chemotherapy groups more likely to Scherwat R, AVLT, ROCFT, ~5 years after chemo of high-dose, 13% of standard- have had mastectomy and higher h 2006 RWT, WAIS-R treatment dose, 3% of non-chemo; no tumor grade. High dose also significant difference among received autologous bone marrow groups. stem cells. High-risk German cohort using German versions of instruments (some modified). FMMSE Cross-sectional; 30 healthy CMF, AC Performance on this task was Long term survivors over the age of Yamada >10 y post- significantly impaired for BC 65. 2010 treatment patients relative to age-matched non-cancer controls Abbreviations: ET: endocrine therapy; RBANS: Repeatable Battery of Adult Neuropsychological Status; WCST: Wisconsin Card Sorting Test; WAIS-III: Wechsler Adult Intelligence Scale-III; GP: Groved Pegboard; DSC: WAIS-III Digit Symbol Coding; WMS-III: Wechsler Memory Scale-III; PASAT: Paced Auditory Serial Addition; CCC:Auditory Consonant Trigrams Test; BNT: Boston Naming Test; COWA: Controlled Oral Word Association ; CVLT: California Verbal Learning Test; WMS-R: Wechsler Memory Scale-Revised; CPT: Continuous Performance Test; RVLT: Rey Visual Learning Test; AVLT: Auditory Verbal Learning Task; CNS-VS: CNS Vital Signs; HLVT-R: Hopkins Verbal Learning Test-Revised, BVMT-R: Brief Visuospatial Memory Test-Revised; TMT: Trail Making Test; TAP: Test for Attentional

5 Performance; ROCFT: Rey-Osterrieth Complex Figure Test; RWT: Regensburg word Fluency Test; FMMSE: Folstein Mini-Mental State Exam; WRAT: Wide Range

Achievement Test. AC: adriamyacin, cyclophosphamide; AC-T: AC+taxotere; CTC: cyclophosphamide, thiotepa, carboplatin; CAF: adriamycin, cyclophosphamide, 5-fluorouracil; CMF: cyclophosphamide, methotrexate, fluorouracil; ACT: adriamyacin, cyclophosphamide, and a taxane; FEC: fluorouracil, epirubicin, cyclophosphamide; FECT: FEC + taxotere; CT: cyclophosphamide, taxotere; DC: docetaxil, cyclophosphamide; taxC: a taxane and carboplatin; GC: gemcitabine, cisplatin; AP: adriamycin, paclitaxel; CFP: cyclophosphamide, 5-fluorouracil, paclitaxel; ET: endocrine therapy; EC: epirubicin, cyclophosphamide; ECT: CTM: cyclophosphamide, thiotepa, mitoxantrone; ACP: adriamyacin, cyclophosphamide, paclitaxel.

Table 3. Motor Function

Instrument Design n per Controls Chemo Conclusion Notes/Limitations Ref group Grooved Longitudinal 30 none AC There were no significant differences in None of women taking ET at time of Jansen Pegboard baseline, ~ 1-4 performance of the dominant or non- testing. 2008 weeks after dominant (although p=0.06) hand between final chemo baseline and after completion of infusion chemotherapy. Grooved Longitudinal 18 None FAC Decline in 1/18 patients relative to baseline Wefel Pegboard baseline, 3 w at 3 weeks post-chemo 2004 after chemo, 1 y later. Grooved Cross-sectional 28 None; CMF Chemotherapy is associated with a Younger women (mean: 42 yo). Wienike Pegboard Within 18 used (majority), significant decline from normative values 1995 months normative CMF + in both dominant (~1 SD) and non- following values CAF, CAF dominant (<1 SD) hands.

7 chemotherapy FePsy Cross-sectional 19-34 BC-no CMF +/- Women treated with chemotherapy had Dutch program and cohort. Schagen Automated ~2 yr after start chemo ET impaired performance on the Finger 1999 Testing of treatment Tapping task (both dominant and non- (finger dominant) relative to patients who did not tapping) receive chemotherapy after surgery Composite Longitudinal 30-31 BC-no ACT Patients not treated with chemotherapy Older post-menopausal women Tager Score baseline, after chemo (majority), showed improvement in performance over (mean age 61). Women who 2010 (Grooved chemo, 6 mo AC, CMF time whereas those treated with withdrew mid-study were more Pegboard, later chemotherapy showed a nonsignificant impaired than the group mean on Finger decline from baseline to post-treatment several objective measures. Tapping) testing approximately 18 months later Diagnosis, ET use varied by group. (p=0.08). ET use also varied by time point. AC: adriamyacin, cyclophosphamide; CAF: adriamycin, cyclophosphamide, 5-fluorouracil; CMF: cyclophosphamide, methotrexate, fluorouracil; ACT: adriamyacin, cyclophosphamide, and a taxane; FAC: fluorouracil, adriamyacin, cyclophosphamide; ET: endocrine therapy.

Table 4. Processing Speed

Instrument Design n per group Controls Chemo Conclusion Notes/Limitations Ref

Letter Longitudinal 43-50 healthy FEC, CMF, Performance impaired in BC relative Composite score based on speed Shilling Cancellation baseline, 6 mo AC to healthy controls; performance and accuracy. 2005 improving in both groups with repeated testing. Letter Longitudinal 43-85 BC-no FEC, CMF, Cognitive deficits only in a few 59% of chemotherapy group were Jenkins Cancellation baseline, 6 mo, chemo, AC, EC, women. node positive versus 14% of non- 2006 18 mo healthy ECP, chemotherapy. ECMF, EFEC TMT-A Longitudinal 18 None FAC Decline in 2/18 patients relative to Wefel baseline, 3 w baseline at 3 weeks post-chemo; 2004 after chemo, 1 y later. TMT-A Longitudinal 60 healthy FEC-T, Chemotherapy impaired performance Canadian cohort. Collins baseline and FEC, CT, relative to controls at 2 of the 7 time 2013 before each AC-T, AC points spanning treatment.

8 treatment TMT-A Longitudinal 28 None CMF 7% experienced a 2 SD decline in Older women (mean: 71 yo). Hurria baseline, 6 mo (majority), performance from baseline to post- Protocol was modified 25% of the 2006 after chemo ACT, AC, chemotherapy testing. way through data collection. AC-T-H TMT-A Longitudinal 101 None EC+P or Similar to norms at both time points Preoperative chemotherapy. Hermelink baseline, 6 mo E+P+CMF German cohort and version of 2007 after chemo instruments. Imperfectly matched normative samples were used to estimate practice effects. TMT-A Longitudinal 40-45 BC-no AC, TAC, Performance improved over time Canadian cohort Diff. in age, Quesnel baseline, chemo, FEC from baseline to 3 months post- tumor stage between cancer 2009 immediately healthy treatment in both chemotherapy and groups. Control group was tested post-chemo, 3 radiotherapy patients. only once, so practice effects could mo later not be determined in this group.

Continued

Table 4 continued Instrument Design n per group Controls Chemo Conclusion Notes/Limitations Ref

TMT-A Longitudinal 12-34 Cardiac CEF Performance improved with repeated Group differences in age, sex, and Mehlsen baseline, 1-2 mo patients, testing. No group differences. education. Danish cohort. 2008 post-chemo healthy controls TMT-A Cross-sectional 28 None; CMF A significant decline in women tested Younger women (mean: 42 yo). Weineke Within 18 used (majority), within 18 months of completing 1995 months normative CMF + chemotherapy relative to normative following values CAF, CAF values; but they remain within 1 SD chemotherapy below normal. TMT-A Cross-sectional; 30 healthy CMF, AC Performance was significantly Long term survivors over the age Yamada >10 y post- impaired in BC survivors relative to of 65. 2010 treatment healthy controls.

TMT-A Cross-sectional 19-34 BC-no CMF +/- ET Women treated with chemotherapy Dutch cohort. Schagen

5 ~2 yr after start chemo exhibited a non-significant trend 1999

9 of treatment toward impaired performance relative

to patients who did not receive chemotherapy after surgery (p=0.08). Symbol Digit Longitudinal 40-45 BC-no AC, AC-T, Radiation patients had lower written Canadian cohort Diff. in age, Quesnel Modalities baseline, chemo, FEC results than healthy controls. A non- tumor stage between cancer 2009 Test immediately healthy significant trend in the same direction groups. Control group was tested post-chemo, 3 existed for the oral portion; no only once, so practice effects could mo later apparent chemotherapy effect. not be determined in this group. Digit Symbol Longitudinal 18 None FAC Decline in 2/18 patients relative to Wefel Coding baseline, 3 w baseline at 3 weeks post-chemo. 2004 (WAIS-R) after chemo, 1 y later. Digit Symbol Cross-sectional 28 None; CMF No significant deviation from Younger women (mean: 42 yo). Wieneke Coding Within 18 used (majority), normative values among women 1995 (WAIS-R) months normative CMF + tested within 18 months of following values CAF, CAF completing chemotherapy chemotherapy Continued

Table 4 continued

Instrument Design n per group Controls Chemo Conclusion Notes/Limitations Ref

DSC Longitudinal 60 healthy FEC-T Significant impairment of Canadian cohort. Collins baseline and (majority), performance among chemotherapy 2013 shortly before FEC, CT, patients relative to healthy controls at each subsequent AC-T, AC. several time points spanning chemo infusion treatment. DSC Cross-sectional 22-35 BC-no 20 diff. Performance was significantly Ahles ~10 years after chemo, regimens impaired for BC and lymphoma 2002 treatment lymphoma patients treated with chemotherapy versus local therapy; DSC Cross-sectional 19-34 BC-no CMF +/- ET Women treated with chemotherapy Dutch cohort. Schagen ~2 yr after start chemo had significantly lower scores than 1999 of treatment patients who did not receive chemotherapy after surgery;

6 DSC&Search Longitudinal 12-34 Cardiac CEF Performance improved with repeated Group differences in age, sex, and Mehlsen

(WAIS-III- baseline, 1-2 mo patients, testing; BC patients performed better education. Danish cohort. 2008 Danish) post-chemo healthy than cardiac patients. controls DSC (WAIS- Longitudinal 101 None ECP or Women being treated for breast Preoperative chemotherapy. Hermelink R German) baseline, 6 mo EPCMF cancer were similar to test norms at German cohort and version of 2007 after chemo both the pre-chemotherapy baseline instruments. Imperfectly matched and near the end of chemotherapy normative samples were used to estimate practice effects. Composite Longitudinal 45-60 BC-no ACP,AC, Older patients with baseline WRAT-3 80% of chemo and 66% of BC-no Ahles Score for baseline, 1 mo, chemo, CAF,FEC, reading scores below the median chemo groups on ET. Sample: 2010 Processing 6 mo, 18 mo healthy CMF treated with chemo had impaired primarily well-educated Caucasian Speed (DSC, post-treatment processing speed women. TMT CWIT, compared to patients not treated with GP) chemotherapy and healthy controls Abbreviations: WAIS: Wechsler Adult Intelligence Scale; DSC: WAIS-III Digit Symbol Coding; TMT: Trail-Making Test: CWIT: Color-Word Inference Test, GP: Grooved Peg Board); WRAT: Wide Range Achievement Test; AC: adriamyacin, cyclophosphamide; AC-T: AC+taxotere; AC-T-H: AC-T+trastuzumab; CTC: cyclophosphamide, thiotepa, carboplatin; CAF: adriamycin, cyclophosphamide, 5-fluorouracil; CMF: cyclophosphamide, methotrexate, fluorouracil; ACT: adriamyacin, cyclophosphamide, and a taxane; FEC: fluorouracil, epirubicin, cyclophosphamide; FECT: FEC + taxotere; CT: cyclophosphamide, taxotere; DC: docetaxil, cyclophosphamide; AP: adriamycin, paclitaxel; CFP: cyclophosphamide, 5-fluorouracil, paclitaxel; ET: endocrine therapy; EC: epirubicin, cyclophosphamide; ECT: CTM: cyclophosphamide, thiotepa, mitoxantrone; ACP: adriamyacin, cyclophosphamide, paclitaxel; EC: epirubicin, cyclophosphamide; ECP: EC+paclitaxel; ECMF: epirubicin+CMF; EPCMF: epirubicin, paclitaxel, CMF; EFEC: epirubicin+FEC. 60

Table 5. Visuospatial Ability and Visual Learning and Memory

Instrument Design n per Controls Chemo Conclusion Notes/Limitations Ref group RCF-IR, Longitudinal 43-85 BC-no FEC, CMF, Cognitive deficits only in a few 59% of chemotherapy group were Jenkins RCF-DR, baseline, 6 mo, 18 chemo, AC,EC,ECP, women. node positive versus 14% of non- 2006 RCF-C mo healthy ECMF,EFEC chemotherapy. RCF-IR, Longitudinal 43-50 healthy FEC, CMF, Chemotherapy had no effect. Composite score based on speed Shilling RCF-DR, baseline, 6 mo AC and accuracy. 2005 RCF-IR, Longitudinal 12-19 BC-no AC+/-ET AC+ET group declined in Very small n and high attrition. Bender RCF-DR, baseline, 1 wk chemo, (majority) performance on both measures from 2006 after completed CMF+/-ET T2 to T3 and was significantly lower chemo, 1 yr later than AC+noET at T3 for RCF-IR RCF-C Longitudinal 30-31 BC-no ACT There was improved performance Older post-menopausal women Tager baseline, after chemo (majority), across testing at baseline, post (mean age 61). Diagnosis and ET 2010 chemo, 6 mo later AC, CMF treatment and 6 months post- use varied by group. ET use also treatment, but no chemo effect varied by time point.

6 RCF-IR, Longitudinal 40-45 BC-no AC, AC-T, Performance on the copy immediate Canadian cohort. Diff. in age, Quesnel

RCF-DR, baseline, chemo, FEC and delayed recalls improved over tumor stage between cancer 2009 RCF-C immediately post- healthy time from baseline to 3 months post- groups. Control group tested only chemo, 3 mo later treatment in both chemotherapy and once, so practice effects could not radiotherapy patients be determined. RCF-IR, Longitudinal 28 None CMF Decline in RCFT Copy, Immediate Older women (mean age 71). Hurria RCF-DR, baseline, 6 mo (majority), Recall, and Delayed Recall from Protocol was modified 25% of the 2006 RCF-C after chemo ACT, AC, baseline to 6 months after way through data collection to AC-T-H chemotherapy. reduce participant burden. RCF-IR, Longitudinal 12-34 Cardiac CEF No sig.change from pre-chemo to 2 Group differences in age, sex, and Mehlsen RCF-DR, baseline, 1-2 mo patients, months post-treatment; no differences education. Danish cohort. 2008 RCF-C post-chemo healthy relative to healthy controls or cardiac RCF-R controls patients RCF-DR, Cross-sectional 28 None; used CMF Significant declines of greater than Younger women (mean: 42 yo). Wieneke RCF-C Within 18 months normative (majority), 1SD from normative values among 1995 following values CMF + CAF, women tested within 18 months of chemotherapy CAF completing chemotherapy RCF-IR, Cross-sectional 19-34 BC-no CMF +/- ET Chemo patients had significantly Dutch cohort. Schagen RCF-DR, ~2 yr after start of chemo lower scores. 1999 RCF-C treatment 61 Continued

Table 5 continued Instrument Design n per Controls Chemo Conclusion Notes/Limitations Ref group RCF-C Cross-sectional; 30 healthy CMF, AC The performance of BC patients was Long term survivors over the age Yamada >10 y post- similar to age-matched non-cancer of 65. 2010 treatment controls. BVMT Longitudinal 60 healthy FEC-T No significant difference between Canadian cohort. Collins baseline and (majority), healthy controls and chemotherapy 2013 shortly before each FEC, CT, patients at any of the 7 time points treatment AC-T, AC. spanning treatment BVRT Cross-sectional; 30 healthy CMF, AC No sig. difference although it Long term survivors over the age Yamada >10 y post-chemo approached significance (p=0.08; of 65. 2010 lower in chemo group). BFRT Cross-sectional; 30 healthy CMF, AC Performancesimilar for BC patients Long term survivors over the age Yamada >10 y post-chemo and age-matched non-cancer controls of 65. 2010

6 VSRT & Longitudinal 18 None FAC Decline in 1/18 patients relative to Wefel

NVSRT baseline, 3 w after baseline at 3 weeks post-chemo. 2004 chemo, 1 y later. Composite Longitudinal 40-53 BC-no FEC (~50%), Decline in chemo group between All participants post-menopausal. Collins score (RVLT baseline, 1 mo chemo CEF, CAF, baseline and immediately after chemo Higher tumor grade in chemo 2009 tasks and after chemo, 1 AC, ACT, no diff. 1 year later. Baseline group. Canadian cohort. At WMS-III year after chemo ECT performance on Family pictures and baseline, chemo groups scored Family RVLT-delayed was higher in chemo higher on several measures. Pictures) group. Composite Longitudinal 30-31 BC-no ACT At baseline, the chemo group had Older post-menopausal women Tager score (BSRT baseline, after chemo (majority), impaired performance relative to the (mean age 61). Diagnosis and ET 2010 and BVRT) chemo, 6 mo later AC, CMF non-chemo group use varied by group, time point Abbreviations: BVRT: Benton Visual Retention Test –Revised; BFRT: Benton Facial Recognition Test; BSRT: Buschke Selective Reading Test; BVMT: Brief Visuospatial Memory Test; RCF-IR: Rey Complex Figure Immediate Recall, RCF-TD: Rey Complex Figure Task Delay; RCF-C Rey Complex Figure Copy; RCF-R: Rey Complex Figure Recognition; VSRT: visuo-spatial reasoning task; RCF-DR: Rey Complex Figure Delayed Recall; VSRT: Verbal Selective Reminding Test; NVSRT: Nonverbal Selective Reminding Test. AC: adriamyacin, cyclophosphamide; AC-T: AC+taxotere; AC-T-H: AC-T+trastuzumab; CTC: cyclophosphamide, thiotepa, carboplatin; CAF: adriamycin, cyclophosphamide, 5-fluorouracil; CMF: cyclophosphamide, methotrexate, fluorouracil; ACT: adriamyacin, cyclophosphamide, and a taxane; FEC: fluorouracil, epirubicin, cyclophosphamide; FECT: FEC + taxotere; CT: cyclophosphamide, taxotere; DC: docetaxil, cyclophosphamide; AP: adriamycin, paclitaxel; CFP: cyclophosphamide, 5- fluorouracil, paclitaxel; ET: endocrine therapy; EC: epirubicin, cyclophosphamide; ECT: CTM: cyclophosphamide, thiotepa, mitoxantrone; ACP: adriamyacin, cyclophosphamide, paclitaxel; EC: epirubicin, cyclophosphamide; ECP: EC+paclitaxel; ECMF: epirubicin+CMF; EPCMF: epirubicin, paclitaxel, CMF; EFEC: epirubicin+FEC.

Table 6. Working Memory and Executive Function

Instrument Design n per Controls Chemo Conclusion Notes/Limitations Ref group Stroop Longitudinal 30 none AC Women treated with chemotherapy None of women taking ET at time Jansen 2008 baseline, 1-4 showed improved performance on of testing. weeks after last this task from baseline to post- chemo treatment testing. Stroop Longitudinal 43-50 healthy FEC, CMF, No group differences, no change Composite score based on speed aShilling baseline, 6 mo AC over time. and accuracy. 2005 Stroop Longitudinal 43-85 BC-no FEC, CMF, No group differences, improved 59% of chemotherapy group were Jenkins 2006 baseline, 6 mo, chemo, AC, EC, performance over time. node positive versus 14% of non- 18 mo healthy ECP, chemotherapy. ECMF, EFEC DS Cross-sectional 28 None; used CMF Chemo group had lower Younger women (mean: 42 yo). Wieneke Within 18 normative (majority), performance relative to normative 1995 months post values CMF + values, but were still within 1 SD

3 chemo CAF, CAF below normal DS Longitudinal 40-45 BC-no AC, AC-T, Baseline performance before Canadian cohort. Diff. in age, Quesnel 2009 baseline, chemo, FEC initiating chemotherapy was tumor stage between cancer immediately healthy reduced relative to healthy groups. Control group tested post-chemo, 3 controls. once; so practice effect mo later undetermined. DS Longitudinal 51-61 BC-no FEC(majorit Freq. of decline was greater in Canadian cohort. Includes only Stewart 2008 baseline, chemo y), CEF, chemo patients than HRT across post-menopausal women. immediately FAC, AC, time. post-chemo ACT, ECT DS Longitudinal 101 None ECP or Women being treated for breast Preoperative chemotherapy. Hermelink baseline, 6 mo EPCMF cancer were similar to test norms at German cohort and version of 2007 after chemo the pre-chemotherapy baseline and instruments. Imperfectly matched better than test norms at the end of normative samples were used to chemotherapy. estimate practice effects. DS Cross-sectional; 30 healthy CMF, AC Performance of BC patients was Long term survivors over the age Yamada 2010 >10 y post- similar to age-matched non-cancer of 65. chemo controls. 63 Continued

Table 6 continued

Instrument Design n per Controls Chemo Conclusion Notes/Limitations Ref group DS/ DSB Cross-sectional 19-34 BC-no CMF +/- ET Chemotherapy was associated with Dutch cohort and instrument. Schagen 1999 ~2 yr after start chemo significantly lower scores on the of treatment DSB; there was no group difference in DS. DSB Longitudinal 43-50 healthy FEC, CMF, Deficit detectable 6 mo after Composite score based on speed Shilling 2005 baseline, 6 mo AC chemo. and accuracy. CCC Longitudinal 51-61 BC-no FEC(majorit Freq. of decline was greater in Canadian cohort. Includes only Stewart 2008 baseline, chemo y), CEF, chemo patients than HRT across post-menopausal women. immediately FAC, AC, time. post-chemo ACT, ECT CCC Longitudinal 60 healthy FEC-T Significant reduction in Canadian cohort. Collins 2013 baseline and (majority), performance relative to healthy before each FEC, CT, controls apparent as early chemo treatment AC-T, AC. following first chemotherapy

6 treatment. 4 WCST Cross-sectional; 30 healthy CMF, AC BC patients committed Long term survivors over the age Yamada 2010

>10 y post- significantly more perseverative of 65. chemo errors than age-matched non- cancer controls. WMS-LNS Longitudinal 43-50 healthy FEC, CMF, There was a significant time*group Composite score based on speed Shilling 2005 baseline, 6 mo AC interaction; controls improved and accuracy. while BC patients did not. WMS-LNS Longitudinal 12-34 Cardiac CEF No group differences, no change Group differences in age, sex, and Mehlsen 2008 baseline, 1-2 mo patients, over time. education. Danish cohort. post-chemo healthy controls WMS-LNS Longitudinal 60 healthy FEC-T No significant difference between Canadian cohort. Collins 2013 baseline and (majority), healthy controls and chemotherapy before each FEC, CT, patients at any of the 7 time points chemo treatment AC-T, AC. spanning treatment WMS-LNS Cross-sectional; 30 healthy CMF, AC BC patients were significantly Long term survivors over the age Yamada 2010 >10 y post- impaired relative to age-matched of 65. No similar difference for chemo non-cancer controls. Digit Span or Arithmetric Score. Continued 64

Table 6 continued

Instrument Design n per Controls Chemo Conclusion Notes/Limitations Ref group TMT-B Longitudinal 101 None ECP or Chemotherapy was associated with Preoperative chemotherapy. Hermelink baseline, 6 mo EPCMF impaired performance on this task German cohort and version of 2007 after chemo relative to test norms at the pre- instruments. Imperfectly matched chemotherapy baseline time point; normative samples were used to there were no post-chemotherapy estimate practice effects. group differences. TMT-B Cross-sectional; 30 healthy CMF, AC Performance on this task was Long term survivors over the age Yamada >10 y post- significantly impaired for BC of 65. 2010 chemo patients relative to age-matched non-cancer controls. TMT-B Cross-sectional 19-34 BC-no CMF +/- ET Women treated with chemotherapy Dutch cohort and instrument. Schagen 1999 ~2 yr after start chemo exhibited a significant impairment of treatment relative to patients who did not receive chemotherapy after

5 surgery. TMT-B Longitudinal 60 healthy FEC-T . No significant difference between Canadian cohort. Collins 2013 baseline and (majority), healthy controls and chemotherapy before each FEC, CT, patients at any of the 7 time points chemo treatment AC-T, AC. spanning treatment. TMT-B Cross-sectional 28 None; used CMF No significant deviation from Younger women (mean: 42 yo). Wieneke Within 18 normative (majority), normative values among women 1995 months post values CMF + tested within 18 months of chemo CAF, CAF completing chemotherapy. TMT-B Longitudinal 40-45 BC-no AC, AC-T, Performance on the copy Canadian cohort. Diff. in age, Quesnel 2009 baseline, chemo, FEC immediate and delayed recalls tumor stage between cancer immediately healthy improved over time from baseline groups. Control group tested post-chemo, 3 to 3 months post-treatment in both once; so practice effect mo later groups. undetermined. I/E Shift20 Cross-sectional; 30 healthy CMF, AC Performance significantly impaired Long term survivors over the age Yamada task >10 y post- in BC patients controls. of 65. 2010 chemo Continued

Table 6 continued Instrument Design n per Controls Chemo Conclusion Notes/Limitations Ref group Composite Longitudinal 40-53 BC-no FEC Summary score decline between All participants post-menopausal. Collins 2009 (WAIS-III baseline, 1 mo chemo (~50%), first two times was greater in Higher tumor grade in chemo tasks and after chemo, 1 CEF, CAF, chemo group; no significant group. Canadian cohort. At WMS-III year after chemo AC, ACT, differences 1 year later. baseline, chemo groups scored SS) ECT higher on several measures. Working Longitudinal 30-31 BC-no ACT There was no apparent treatment or Study conducted in older post- Tager 2010 Memory & baseline, after chemo (majority), time effect at baseline, post menopausal women (mean age Attention chemo, 6 mo AC, CMF treatment or 6 months post- 61). Women who withdrew mid- Composite later treatment among breast cancer study were more cognitively (TMT, patients who received or did not impaired than the group mean on DS,NLS, receive chemotherapy. several objective measures. Arithmetic) Diagnosis and ET use varied by group. ET use also varied by time point.

6 Abbreviations: WCST: Wisconsin Card Sorting Test; DS: Wais-III Digit Span; DSB: Digit Span Backwards; WMS-LNS: Wechsler Memory Scale-Letter Number Span;

CCC: Consonant Trigrams. TMT-B: Trail Making Test B; I/E Shift 20 Task: Intradimensional/ Extradimensional Shift 20 Task AC: adriamyacin, cyclophosphamide; AC-T: AC+taxotere; AC-T-H: AC-T+trastuzumab; CTC: cyclophosphamide, thiotepa, carboplatin; CAF: adriamycin, cyclophosphamide, 5-fluorouracil; CMF: cyclophosphamide, methotrexate, fluorouracil; ACT: adriamyacin, cyclophosphamide, and a taxane; FEC: fluorouracil, epirubicin, cyclophosphamide; FECT: FEC + taxotere; CT: cyclophosphamide, taxotere; DC: docetaxil, cyclophosphamide; AP: adriamycin, paclitaxel; CFP: cyclophosphamide, 5- fluorouracil, paclitaxel; ET: endocrine therapy; EC: epirubicin, cyclophosphamide; ECT: CTM: cyclophosphamide, thiotepa, mitoxantrone; ACP: adriamyacin, cyclophosphamide, paclitaxel; EC: epirubicin, cyclophosphamide; ECP: EC+paclitaxel; ECMF: epirubicin+CMF; EPCMF: epirubicin, paclitaxel, CMF; EFEC: epirubicin+FEC.

Table 7. Verbal Ability, Learning, and Memory

Instrument Design n per Controls Chemo Conclusion Notes/Limitations Ref group HVLT-R Longitudinal 60 healthy FEC-T, FEC, Difference in total score between Canadian cohort. Collins 2013 baseline and before CT, AC-T, AC groups at 1/7 time points during each treatment treatment. AVLT Longitudinal 43-50 healthy FEC, CMF, AC A significant time*group interaction Composite score based on Shilling baseline, 6 mo existed for both Supraspan and Total speed and accuracy. 2005 recall: controls improved while BC patients declined. AVLT Longitudinal 43-85 BC-no FEC, CMF, AC, No significant time or group effects. 59% of chemotherapy group Jenkins 2006 baseline, 6 mo, 18 chemo, EC, ECP, were node positive versus mo healthy ECMF, EFEC 14% of non-chemotherapy. AVLT Cross-sectional 19-34 BC-no CMF +/- ET Chemotherapy was associated with an Dutch cohort. Schagen ~2 yr after start of chemo impairment in delayed recall, but not 1999 treatment immediate recall or recognition,

6 relative to patients who did not receive

chemotherapy after surgery (p=0.08) AVLT Cross-sectional; 30 healthy CMF, AC BC patients committed significantly Long term survivors over the Yamada >10 y post-chemo more perseverative errors than age- age of 65. 2010 matched non-cancer controls. AVLT Longitudinal 12-34 Cardiac CEF No significant time or group effects. Group differences in age, Mehlsen baseline, 1-2 mo patients, sex, and education. Danish 2008 post-chemo healthy cohort. controls CVLT Longitudinal 40-53 BC-no FEC (~50%), Baseline performance was significantly All post-menopausal. Higher Collins 2009 baseline, 1 mo after chemo CEF, CAF, AC, better among patients who tumor grade in chemo group. chemo, 1 year after ACT, ECT subsequently received chemotherapy Canadian cohort. At chemo versus those who had hormone therapy baseline, chemo groups without chemotherapy. scored higher on several measures. CVLT Cross-sectional 28 None; CMF A significant decline from normative Younger women (mean: 42 Wieneke Within 18 months used (majority), values in the Long Delay, but still yo). 1995 following normative CMF + CAF, within 1 SD of normal. No effect on chemotherapy values CAF CVLT 1-5 or Short Delay Tests. Continued 67

Table 7 continued

Instrument Design n per Controls Chemo Conclusion Notes/Limitations Ref group CVLT Cross-sectional 22-35 BC-no 20 diff. Scores were comparable for BC and Ahles 2002 ~10 years after chemo, regimens lymphoma patients treated with treatment lymphoma chemotherapy versus local therapy nearly a decade after treatment. BNT Cross-sectional; 30 healthy CMF, AC Performance was similar for BC Long term survivors over the Yamada >10 y post-chemo patients and age-matched non-cancer age of 65. 2010 controls. COWAT Longitudinal 60 healthy FEC-T, FEC, Significant impairment in Canadian cohort. Collins 2013 baseline and before CT, AC-T, AC chemotherapy patients emerge at later each treatment time points spanning treatment. Composite: Longitudinal 30-31 BC-no ACT (majority), There was improved performance Older post-menopausal Tager 2010 BNT baseline, after chemo AC, CMF across testing at baseline, post women (mean age 61). +COWAT chemo, 6 mo later treatment and 6 months post-treatment, Diagnosis and ET use varied

8 but no difference between breast cancer by group. ET use also varied patients who received or did not by time point receive chemotherapy. VFT Longitudinal 40-45 BC-no AC, AC-T, FEC Performance on the task decreased Canadian cohort. Diff. in Quesnel baseline, chemo, from baseline to the post-treatment age, tumor stage between 2009 immediately post- healthy time point among chemotherapy cancer groups. Control group chemo, 3 mo later patients but not radiotherapy patients tested once; so practice effect undetermined. 4WSMT Longitudinal 12-19 BC-no AC+/-ET Chemo +/-ET did worse on the 5-sec, Very small n and high Bender 2006 baseline, 1 wk after chemo, (majority) 15-sec and 30-sec tasks than non- attrition. chemo, 1 yr later CMF+/-ET chemo group when tested approximately 1 year after chemo. WMS-LM Longitudinal 43-50 healthy FEC, CMF, AC Baseline difference existed between Composite score based on Shilling baseline, 6 mo BC and controls prior to chemo. speed and accuracy. 2005 WMS-LM Longitudinal 43-85 BC-no FEC, CMF, AC, No significant time or group effects. 59% of chemotherapy group Jenkins 2006 baseline, 6 mo, 18 chemo, EC, ECP, were node positive versus mo healthy ECMF, EFEC 14% of non-chemotherapy. Continued

Table 7 continued

Instrument Design n per Controls Chemo Conclusion Notes/Limitations Ref group WMS-LM Longitudinal 40-53 BC-no FEC (~50%), Baseline performance was significantly All post-menopausal. Higher Collins 2009 baseline, 1 mo after chemo CEF, CAF, AC, better among patients who tumor grade in chemo group. chemo, 1 year after ACT, ECT subsequently received chemotherapy Canadian cohort. At chemo versus those who hormone therapy baseline, chemo groups without chemotherapy. scored higher on several measures. WMS-LM Longitudinal 12-34 Cardiac CEF No significant time or group effects. Group differences in age, Mehlsen baseline, 1-2 mo patients, sex, and education. Danish 2008 post-chemo healthy cohort. controls WMR-LM Cross-sectional 22-35 BC-no 20 diff. Scores in logical memory I and II were Ahles 2002 ~10 years after chemo, regimens lower for breast cancer and lymphoma treatment lymphoma patients treated with chemotherapy versus local therapy.

9 WMR-LM Longitudinal 101 None ECP or EPCMF Women being treated for breast cancer Preoperative chemotherapy. Hermelink

baseline, 6 mo after were similar to test norms at the pre- German cohort and version 2007 chemo chemotherapy baseline and better than of instruments. Imperfectly test norms at the end of chemotherapy. matched normative samples were used to estimate practice effects. Composite Longitudinal 45-60 BC-no ACP,AC, Chemotherapy had an acute effect on 80% of chemo and 66% of Ahles 2010 Score for baseline, 1 mo, 6 chemo, CAF,FEC, verbal ability that was detected upon BC-no chemo groups on ET. Verbal Ability mo, 18 mo healthy CMF completion of treatment and then Sample: primarily well- (WASI, D- resolved over the subsequent 18 educated Caucasian women. DEFS) months. Abbreviations: HVLT-R: Hopkins Verbal Learning Test-Revised. WMS-LM: Wechsler Memory Scale-Logical Memory. AVLT: Rey Auditory-verbal Learning Test; CVLT: California Verbal Learning Test. COWAT: Controlled Oral Word Association Test; VFT: Verbal Fluency Test. 4WSMT; 4 Word Short Memory Test; BNT: Boston Naming Test; D-KFES: Delis-Kaplan Executive Function System. AC: adriamyacin, cyclophosphamide; AC-T: AC+taxotere; AC-T-H: AC-T+trastuzumab; CTC: cyclophosphamide, thiotepa, carboplatin; CAF: adriamycin, cyclophosphamide, 5-fluorouracil; CMF: cyclophosphamide, methotrexate, fluorouracil; ACT: adriamyacin, cyclophosphamide, and a taxane; FEC: fluorouracil, epirubicin, cyclophosphamide; FECT: FEC + taxotere; CT: cyclophosphamide, taxotere;; AP: adriamycin, paclitaxel; CFP: cyclophosphamide, 5-fluorouracil, paclitaxel; ET: endocrine therapy; EC: epirubicin, cyclophosphamide; ECT: CTM: cyclophosphamide, thiotepa, mitoxantrone; ACP: adriamyacin, cyclophosphamide, paclitaxel; EC: epirubicin, cyclophosphamide; ECP: EC+paclitaxel; ECMF: epirubicin+CMF; EPCMF: epirubicin, paclitaxel, CMF; EFEC: epirubicin+FEC.

Table 8.Effects of Chemotherapy on Affect in Breast Cancer Patients (Women)

Study Type Instrument Drug Population Testing Time Effect Reference Longitudinal HADS FEC (6-7 cycles) n=382 chemo (42%) or non-chemo 4,6,10 & 16months Twice as many chemo patients Björneklett RT (76.7%), tamoxifen (65.2%); post-treatment had high anxiety scores; no 2012 premenopausal (24.9) difference in depression. mean age: 55,62 Longitudinal IDS-SR Not stated n=64 ; 24 chemo, 40 non-chemo After chemo, during Chemo, not radiation, associated Torres 2013 PSS mean age: 52, 59 post-chemo RT and 6 w/ depression, IL-6, NF-kB, RT: 100% week post RT TNFaR2 and IL-6 Longitudinal HADS Not stated n=220 Baseline (within 9mo No difference in anxiety or Kwiatkowski 75% HRT; RT: 88% post chemo) depression scores across time 2013 Mean age: 52 6,12 mo post baseline Longitudinal PHQ-9 CMF n=32 Weekly over 5 mos; Dep. associated with AC chemo Reece 2013 GAD-7 (6),Taxol(15) mean age 52.4 after chemo started not others; chemo duration, 1st (both based AC (21) premenopausal (5.7%) (90%) chemo regimen, anxiety

0 on DSM IV) Abraxane (7), Other (5) Longitudinal HADS FEC, FEC+T, n=138 chemo, 21 non-chemo Pre-chemo and 1 month Anxiety, not dep., associated with Vearncombe FEA, CAF, CA, Mean age: 49, 54 post-chemo cognition across time. Cog. assoc. 2009 CA+T, CA+Taxol, with baseline dep. and fatigue. CEA, CMF, C+T Cross- POMS Not reported n=175 women, n=88 men While on chemo Depression, anxiety, fatigue, Redecker sectional Mean age: 57 treatment insomnia all related. 2000 17.5% BC; many other cancer Depression is the best predictor types also included of QoL Cross- BDI- 1A Doxorubicin or n= 653 women in 4 age groups: Within 8 mo of Dox treatment increase Avis 2012 sectional other 25-44, 45-54, 55-64, 65-74; RT: diagnosis depression, others did not 25% Cross- BDI Not listed n=534; chemo and non-chemo Within 1 year post Higher depression, lower QoL in Hwang 2013 sectional Mean age: 48,49 respectively chemo, 1-3 years, or 3+ groups <1year and 3+ yrs since HRT(%) 69,78; RT(%): 56,51 years chemo Premenopausal (%): 33,36 Compared to non-chemo Continued

Table 8 continued Study Type Instrument Drug Population Testing Time Effect Reference Cross- HADS-A, Not listed n=72; Metastatic BC Right before receiving high IL-6 associated w/ Jehn 2012 sectional HADS-D mean age: 60 one of their depression, not anxiety. previous chemo (75%) chemotherapy Depression and anx. correlated previous HRT(25%) treatments Cross- BDI-II Not listed n=103; post-chemo (48%), non- Within 3mo post- Fatigue, sleep disturbance, Bower 2011 sectional chemo (52%) treatment depression, and sTNF-rII higher Mean age: 51 in chemo group Cross- POMS-SF AC (6) n=46 1mo post chemo Depression, but not cognition Reid-Arndt sectional Mean age: 53.38 predicted poor QoL 2009 Cross- BDI-II 30% CMF, 23% n=200 While on chemo Treatment duration (months) and Park 2013 sectional FACT-B AC, 20.5% AT, mean age :45 years # of chemo cycles associated with 26.5% other menopausal symptoms and sexual dysfunction, but not depression Cross- HADS Not listed n=218 chemo (60%) or RT (40%) Midway through Anxiety and depression were So 2010 sectional Mean age: 51, 53 treatments higher in chemo patients Abbreviations: RT: radiotherapy; HRT: hormone replacement therapy; HADS: Hospital Anxiety and Depression Scale; IDS-SR: Inventory of Depressive Symptomotology

7 Self-Report; PSS: Perceived Stress Scale; PHQ-9: Patient Health Questionnaire 9 item; GAD-7: Generalized Anxiety 7 item Questionnaire; POMS: Profile of Mood States;

BDI: Beck Depression Inventory; POMS-SF: Profile of Mood States Short Form; FACT-B: Functional Assessment of Cancer-Breast. AC: adriamyacin, cyclophosphamide; CAF: adriamycin, cyclophosphamide, 5-fluorouracil; CMF: cyclophosphamide, methotrexate, fluorouracil; ACT: adriamyacin, cyclophosphamide, and a taxane; FEC: fluorouracil, epirubicin, cyclophosphamide; DC: docetaxil, cyclophosphamide; AP: adriamycin, paclitaxel; CFP: cyclophosphamide, 5-fluorouracil, paclitaxel.

Table 9. Effects of chemotherapy on cognition in rodents

Drug and dose Route, duration, frequency Animals Cognitive test (timing) Cognitive outcome Reference (age if available) Doxorubicin (2.5mg/kg) + I.P. each week for 4 weeks Female Sprague-Dawley rats Passive avoidance (2 days Impaired learning Konat 2008 cyclophosphamide (25 (10mo.) after chemo) mg/kg) Young-cyclo(100mg/kg) or I.P. every 4 weeks for 18 weeks Female Fischer rats MWM + Stone 14-unit Increased short term Lee 2006 5-FU (150 mg/kg) (young) or 16 weeks (old) young (7mo.) and T-maze (young: 7 or 29 wks; learning, no difference Old- cyclo(80mg/kg old (18 mo.) old: 16 weeks post chemo) long term Cyclo (40mg/kg) + I.V. (tail) 1/week for 3 weeks Female ovariectomized Cued and contextual fear Impaired contextual, but Macleod dox (4mg/kg) Sprague-Dawley rats (eight Conditioning (1 week post not cued, fear memory 2007 weeks old) chemo) Dox (.5, 2, or 8 mg/kg) Single I.P. inj. Male Wistar rats (180–350 Inhibitory avoidance Impairment of memory Liedke g) conditioning (3h, 24 h, 7 retention 2009 days) Cyclophosphamide Single I.P. injection Male CF1 mice (70–90 days Step-down inhibitory Impaired inhibitory Reiriz 2006 (8, 40, or 200 mg/kg) old) avoidance (1, 7 days) avoidance for 40 and 200

2 mg/kg Cyclophosphamide Single I.P. injection Male ICR mice (8–10 weeks Passive avoidance + NOR Impaired passive Yang 2010 (40mg/kg) old) (12 hrs. post injection) avoidance learning Impaired NOR Cyclo (30 mg/kg) 7 inj. (I.V. tail vein) every other Male Lister-hooded rats NLR (5 days post chemo) No diff. (working Lyons 2011 day memory) Dox (4mg/kg) OR 5-FU I.P. inj. 1/wk for 3 weeks Female C57b./6J (9wks) Context and cued fear No difference Fremouw (100mg/kg) OR Dox and conditioning; NOR (1-2 2012 cyclo (4 and 80 mg/kg) weeks after last treatment) Cyclo and 5-FU Three I.P. inj. 1 month apart Male F344 rats (5 mo.) MWM, Fear conditioning, No impairment Long 2011 Low dose: 50, 75 mg/kg and 14-unit T maze (before High dose: 75, 120 mg/kg and right after last treatment) 5-FU (25 mg/kg) I.V. (tail) every other day for 5 Male Lister-hooded rats Fear conditioning + NLR Impairment in both Elbeltagy days (150–170 g) (day after treatment) 2010 5-FU (20 mg/kg) I.V. (tail vein) inj. 5 times across Male Lister-hooded rats NLR Impairment in NLR Mustafa 12 days (200–250 g) 2008 MTX + 5- FU; High dose: I.P. 1/wk for 4 weeks Male C57BL/6Hsd mice (7-8 Contextual fear No memory impairment Gandal 37.5 MTX, 75 mg/kg 5- FU weeks old) conditioning + NOR (6 2008 Low dose: 38, 19 mg/kg weeks post chemo) 72 Continued Table 9 continued Drug and dose Route, duration, frequency Animals Cognitive test (timing) Cognitive outcome Reference (age if available) MTX (37.5 mg/kg) + 5-FU I.P. inj. 1/week for 3 weeks Female BALB/C mice MWM, NMTS, dNMTS; 1 Impairment in spatial Winocur (75 mg/kg) (2 mo.) month after injection (1 wk memory but not cued 2006 post chemo) memory or discrimination learning MTX (50mg/kg) + 5-FU I.P. inj. 1/wk for 4 weeks Female balb/C (3 mo.) MWM, NMTS and dNMTS Impairment in all tasks but Winocur (75 mg/kg) tasks (1 wk post chemo) cued memory 2011 MTX (37.5 mg/kg) + 5-FU Single I.P. inj. Female balb/C mice (5 mo.) MWM, NMTS and DNMTS Impairment in all tasks but Wincur (50mg/kg) tasks (1 wk post chemo) cued memory 2012 Methotrexate (MTX) Single I.P. inj. Male hooded Wistar rats (8 MWM + NOR + go/no-go Impairment in all 3 tasks Fardell (250mg/kg) mo.) task (week 2 out to 8 mo.) (visual, spatial, and 2010 discrimination learning) MTX (adults 250 mg/kg; Adults: single I.P. Male Long-Evans rats (12 NOR + NLR (3-7 days post Impaired NLR, not NOR Li 2010 young 1mg/kg) Young: 2/wk 1st two weeks, weeks old) and young female chemo) 1/wk for following 6 weeks and male (2 wks old) MTX (1.5-6 mg/kg) 3 injections (1/day) in lateral Male Wistar rats (four Conditioned avoidance Impaired learning and Madhyastha

7 ventricle months old) Test (1 wk post chemo) memory 2002

MTX (250 mg/kg) Sing tail vein injection Male Wistar rats (three MWM + NOR + contextual Impairment in NOR, not Seigers months old) fear conditioning (2 wks) MWM 2009 MTX (0.005 mg/kg) Single I.P. injection Male and female Sprague- Appetitive Pavlovian No impairment Stock 1995 Dawley rats (17 days old). discrimination + conditioned taste aversion (at 80 days old) Paclitaxel (1mg/kg) unclear Male Long-Evans rats 5 choice serial reaction time No impairment (1-19 days Boyette- task post chemo) Davis 2009 Oxaliplatin (8 or 12 mg/kg) Single I.P. inj. Male hooded Wistar rats NOR, MWM, Fear Impairment in NOR, Fardell OR 75 mg/kg 5-FU OR conditioning, context testing MWM, and context- all 2012 combination of both drugs hippocampal related (2-3 wks post chemo) thioTEPA (10mg/kg) I.P. daily for 3 days Male C57BL/6J mice (ﬁve NOR + NLR 1-30 wks post Impairment in both at 12 Mondie weeks old) inj. and 20 weeks 2010 Cytosine arabinoside I.P. injection every day for 5 Male Sprague-Dawley rats MWM 7 days after last Impairment in remote Li 2008 (400 mg/kg) days injection recall of MWM (1 wk post chemo) Abbreviations: I.P.: Intraperitoneal; I.V.: Intravenous; MWM: Morris Water Maze; NOR: Novel Object Recognition; NLR: Novel Location Recognition; NMTS: non- matching to sample; DNMTS: delayed non-matching to sample; MTX: methotrexate. 73

Table 10. Effects of chemotherapy on affective behavior in rodents

Drug and dose Route, duration, Animals Affect test Affective outcome Reference frequency Doxorubicin (7 mg/kg) Single I.P. inj. Male Wistar rats EPM and OF (1, 24,48,and 73 Increased anxiety-like Merzoug hours after inj.) behavior, but decreased 2011 locomotor activity; myelosupression and increase in brain oxidative stress Doxorubicin (9mg/kg) OR Single dose I.V. (tail vein) Pregnant C57bl/6 females OF, passive avoidance, social Increased anxiety (less time Van Vinblastine (6mg/kg) injected, but offspring tested at exploration, EPM (age 12-14 in open arms) Calsteren 3 mo. weeks) 2009 Doxorubicin (10 mg/kg) Single I.V. dose Male Swiss albino mice EPM and Vogel’s conflict test Increased anxiety-like Anwar

7 (8 and 14 days post chemo) behavior, reversed by 2011 4 alprazolam

Doxorubicin (10 mg/kg) Single I.P. inj. Female C57bl/6 (2 mo.) TST and FST, EPM Increased depressive-like Kujjo 2011 (FST, not TST), but not anxiety-like behavior thioTEPA (10mg/kg) I.P. daily for 3 days Male C57BL/6J mice (ﬁve FST 3,6,8,13, and 21 weeks No effect on depressive Mondie weeks old) after inj. behavior 2010 TST 8,13, and 21 weeks after inj. Cyclophosphamide Single I.P. injection Male CF1 mice (70–90 days OF (1, 7 days post chemo) No effect on anxiety Reiriz 2006 (8, 40, or 200 mg/kg) old) behavior Methotrexate + 5- FU I.P. 1/wk for 4 weeks Male C57BL/6Hsd mice (7-8 Contextual fear conditioning Increased anxiety-like Gandal High dose: 37.5 mg/kg weeks old) (6 weeks post chemo) behavior (increased freezing) 2008 MTX, 75 mg/kg 5- FU Low dose: 38, 19 mg/kg MTX (1.5-6 mg/kg) 3 injections (1/day) in Male Wistar rats (four months Light/dark box (1 week post No light/dark diff., but Madhyastha lateral ventricle old) chemo) decreased exploratory 2002 behavior in dark Reduced brain monoamines Abbreviations: MTX: methotrexate; EPM: elevated plus maze; OF: open field test; FST: forced swim test; TST: tail suspension test.

Table 11. Chemo induced inflammatory changes in clinical studies

Study Type Drug (cycles) Population Cytokine Neuropsychological/ Chemo effect Reference assessment Behavioral assessment Cross- Not specified n=103; post- Baseline# Self report measures: FSI Chemo group had higher sTNF-rII, but not IL-1ra or Bower et al sectional chemo (48%), (fatigue), BDI-II CRP; 2011 non-chemo (52%) (depression), and PSQI sTNF-rII associated with fatigue in chemo group. Sleep Mean age: 51 (sleep disturbance) disturbance and depression not associated with inflammatory markers Cross- AC, AP, CFP, n=42 chemo BC 4.8±3.4 HVLTR, WAIS-4 Chemo treated group had higher levels of TNF-a, IL-6, Kessler sectional or CMF patients (range 1-12) Self-report: MMQ memory impairment, and reduced left hippocampal 2013 n= 35 healthy years after (memory) volume. These were all correlated with each other in controls treatment chemo treated group, but not healthy controls. Longitudinal AC/CAF BC patients Prior to cycle Self report (cognition): 5 IL-6 higher in AC/CAF group at 2nd timepoint than Janelsisns (n=27) Or CMF n=27/group; 2 and after questions from FSCL CMF; 2012 (n=27) Mean age: 53, 51 cycle 4 IL-6 increased across time in AC/CAF, but not CMF,

7 Paxil group

(~50%/group) MCP-1 negatively corr. with self-report measures of thinking, concentration, and forgetfulness in AC/CAF but not CMF group

Longitudinal Not specified n=33; BC patients Baseline# Self report (cognition): Higher levels of circulating IL-1ra, sTNF-rII, and C Pomykala Chemo n=23; and 1 year PAOFI reactive protein (CRP) at first time point in chemo 2013 non-chemo n=10 later. group; higher CRP 1 year later Mean age: 51, 56 Positive correlations across time between cytokines (IL- 1ra, and sTNF-RII) and metabolism in medial frontal and temporal cortex No data reported relating cytokines to cognition Longitudinal CMF or CAF BC patients Baseline₤ N/A At baseline, chemo groups had higher TNF-a and IL-6, Tsavaris with Paclitaxel N=15/group and after last and lower IL-2, IFN-y, and GM-CSF than healthy 2002 or Docetaxel except for health cycle controls. Across time, chemo increased IL-6, GM-CSF, controls (n=20) and IFN-y and decreased IL-1, TNF, and PGE2 levels. Change was greater for Paclitaxel than Docetaxel Continued

Table 11 continued Study Type Drug (cycles) Population Cytokine Neuropsychological/ Chemo effect Reference assessment Behavioral assessment Longitudinal Paclitaxel or N=90 BC; 70 Baseline₤, N/A No differences at baseline. Across time IL-6, 8, and 10 Pusztai CAF Paclitaxel, 20 day 3, and was higher in Paclitaxel group than control. No diff. 2004 FAC; after last between FAC and control groups 15 healthy treatment controls Median age: 47 Longitudinal ACT (29%), BC patients Baseline#, WTAR, CVLT-2, WMS- TNF-RII higher in chemo patients at baseline and 6 Ganz 2013 FEC (4%), DC Chemo; n=49 6 months, 12 3, BVMT-R, ROCF, months, but not 1 year; TNF-RII associated with (51%), Non-chemo; n=44 months after WAIS-3, TMT, Stroop, memory complaints in chemo, but not non-chemo taxC(18%), or Mean age: 50, 53 baseline GP patients. TNF-RII is higher in chemo group and is GC (2%) Self-report: FSI (fatigue), associated with diminished metabolism in inferior BDI-II (depression), frontal cortex. PSIQ (sleep), SMQ No relationship between baseline cytokines and

7 (memory) cognitive functioning.

6 No chemo effect on cognitive functioning at baseline;

chemo had lower psychomotor functioning, but not significant (p=.07) Longitudinal Not specified BC patients 1 week Self-report: IDS-SR Chemotherapy was associated with higher levels of NF- Torres Chemo; n=24 before, week (depression), MFI kB DNA binding in peripheral blood mononuclear cells, 2013 Non-chemo; n=40 6 of , and 6 (fatigue), PSS (distress) IL-6, and sTNF-R2 as well as depression and fatigue. Mean age: 59,52 weeks after Chemotherapy and NF-kB DNA binding were sig. RT; Chemo predictors of depression. received before RT. #post-chemotherapy; ₤ post surgery and pre-chemo; §pre-surgery and pre-chemo AC: adriamyacin, cyclophosphamide; CAF: adriamycin, cyclophosphamide, 5-fluorouracil; CMF: cyclophosphamide, methotrexate, fluorouracil; ACT: adriamyacin, cyclophosphamide, and a taxane; FEC: fluorouracil, epirubicin, cyclophosphamide; DC: docetaxil, cyclophosphamide; taxC: a taxane and carboplatin; GC: gemcitabine, cisplatin; AP: adriamycin, paclitaxel; CFP: cyclophosphamide, 5-fluorouracil, paclitaxel

Table 12. Chemo induced changes in brain structure

Study Type Imaging Drug Population Cognitive Tests Testing Period Chemo effect Reference Technique Cross- MRI AC CMF, EC n=106; post-chemo WMS-R Baseline# Chemo group had smaller gray and white Inagaki 2007 sectional PTX, UFT (51), non-chemo (55) matter in prefrontal, parahippocampal, and n= 55 healthy controls cingulated gyrus. These correlated with Mean age: 46 attention, concentration, and visual memory Cross- MRI AC CMF, EC n=132; post-chemo WMS-R Baseline# No difference Inagaki 2007 sectional PTX, 5-FU, (73), non-chemo (59) DFUR, n= 37 healthy controls HCFU, UFT Mean age: 48 Cross- MRI FEC + CTC n=32; post-chemo (17), not completely 9+ years after Chemo treated patients showed reduced de Ruiter, sectional non-chemo (15) specified chemo white matter integrity and gray matter 2011 Mean age: 57, 58 volume Cross- MRI AC, n= 10 chemo Digit Symbol Average 22 Chemo patients had lower white matter Abraham sectional AC+taxane n= 9 healthy controls test months post integrity in the genu of the CC, associated 2008

7 chemotherapy with slower processing speed Cross- MRI CMF, AC, n=75; chemo (44), non- WMS-R More than 3 No difference in hippocampal volume or Yoshikaw20 sectional CAF, CPA, chemo (31) years post memory ability 05 MF, DFUR, Mean age: 48,48 chemo CAR Cross- MRI FEC, n=17 post-chemo BWD, TEA, 80-160 days Chemo group had reduced white matter Deprez 2011 sectional FEC+Taxol n= 18 healthy controls WAIS, AVLT, post chemo integrity which correlated with attention Mean age: 57, 58 RVDLT, and processing/psychomotor speed SCWT, COWA, 9HPT, TMT; Self report: CFQ Cross- MRI AC, AT, CFP, n=42 post-chemo WAIS4 Average 5 Chemo group had reduced left Kesler 2013 sectional CMF n= 35 healthy controls years post hippocampal volume which was associated Mean age: 55, 56 chemo (range with verbal memory performance. 1-12) Cross- MRI CMF n= 184 chemo - Average 21 Chemotherapy treated patients had smaller Koppelmans sectional n= 368 healthy controls years post total brain volume and gray matter 2012 treatment volume; no difference in hippocampal volume or white matter. 77 Continued

Table 12 continued Study Type Imaging Drug Population Cognitive Tests Testing Period Chemo effect Reference Technique Longitudinal MRI AC, ACT, n=28; chemo (27), non- Self-report: Baseline₤ and Chemotherapy treated patients, but not McDonald taxol+carbopl chemo (28) BRIEF 1 month post non-chemo or healthy controls, had 2013 atin, taxol + n= 24 healthy controls treatment decreased frontal gray matter after cisplatin Mean age: 50,52,47 treatment which was associated with executive functioning difficulties Longitudinal MRI AC, n=106; chemo (17), - Baseline₤ No difference at baseline. Chemo had McDonald AC+taxol non-chemo (12) decreased gray matter density at 1 and 12 2010 n= 18 healthy controls months after treatment. No difference in Mean age: 46 non-chemo or control groups. Longitudinal MRI Unspecified n=50; chemo (34), non- not completely Baseline₤ and Chemo patients had decreased white Deprez 2012 chemo (16) specified 3-4 months matter integrity across time; non-chemo n= 19 healthy controls post treatment and healthy controls did not. Attention and Mean age: 44,43,44 verbal ability correlated with white matter integrity changes. Cross- fMRI AC, ACT, n=28; chemo (25), non- WCST, WAIS4 Average 5 Chemo patients had reduced activation in Kesler 2011

8 sectional CarT, CT, chemo (19) digit span, years post left caudal lateral prefrontal cortex FECT, CMF, n= 18 healthy controls DKEFS, treatment compared to non-chemo and healthy ACF Mean age: 56,58,56 Self-report: controls. This was sig. correlated with BRIEF executive dysfunction in the chemo group. Cross- fMRI CMF, ACT n = 14 post-chemo Verbal memory Average 3 During memory encoding, chemotherapy Kessler 2009 sectional n= 14 healthy controls encoding and years post patients had lower activation than controls Mean age: 55,54 recall tasks treatment and CMF treated patients had lower prefrontal cortex activation than ACT and controls. Chemo had higher activation in multiple brain regions during memory compared to healthy controls. Cross- fMRI FEC + CTC n=34; chemo (19), non- TMT, WAIS-III 9 years after Chemo group showed less activity in the de Ruiter sectional chemo (15) DSCT, Stroop, chemo dorsolateral prefrontal cortex during 2011 Mean age: 56, 58 CVLT, WMS-R executive functioning task, VRT, FFTT, parahippocampal gyrus during the memory WFT, tower of task, and bilateral posterior parietal cortex London, paired during both tasks. Chemo patients showed associates task cognitive deficits in verbal memory and planning

Table 12 continued Study Type Imaging Drug Population Cognitive Tests Testing Period Chemo effect Reference Technique Longitudinal fMRI FECT, n=21 post-chemo Verbal recall Baseline₤ and Chemo group showed less anterior Zunini 2012 FECTH, CT, n= 21 healthy controls task 1 month post cingulated activation at baseline than AC Mean age: 50, 50 treatment healthy controls. Across time chemo patients had less activation in multiple areas and had less activation than controls; no change across time in controls Longitudinal fMRI AC, ACT n=28; chemo (16), non- n-back task Baseline₤, 1 Both cancer groups showed increased McDonald, chemo (12) month and 1 bifrontal and decreased left parietal Conroy, n= 15 healthy controls year post activation at baseline and decreased frontal Ahles 2012 Mean age: 53,53,51 treatment hyperactivation 1 month post treatment compared to healthy controls. Chemo effect on frontal lobe 1 month after chemo. Cross- qEEG FEC or FEC+ n=47; high dose chemo Not specified Average 1.5-2 Asymmetry of alpha rhythm in 41% of Schagen sectional high dose (17), std. dose (16), but gave a ref. years post high dose and 12% of standard dose 2001

9 CTC non-chemo (14) chemo patients compared to 0% in non-chemo Mean age: 46,49,49 controls. Cross- PET Not specified n=16 post-chemo 21 5-10 yrs post Chemo group showed altered cerebral Silverman sectional n= 8 controls; non- neuropsychologi chemo blood flow during short term memory 2007 chemo (5), healthy (3) cal tests (not recall task relative to control group. Mean age: 50, 53 specified but Metabolism in basal ganglia was decreased gave a ref.) in tamoxifen chemo group relative to chemo w/out tamoxifen or controls Case study MRI and ACT n=2 (monozygotic n-back task 22 months post Chemo patient had increased white matter Ferguson fMRI twins); chemo (1), non- chemo (WM) lesions in both cerebral 2007 chemo (1) hemispheres and increased brain activity Age: 60 during working memory task; No WM volume differences in other brain regions. #post-chemotherapy; ₤ post surgery and pre-chemo; §pre-surgery and pre-chemo AC: adriamyacin, cyclophosphamide; CMF: cyclophosphamide, methotrexate, fluorouracil; EC: epirubicin, cyclophosphamide PTX: paclitaxel; UFT: tegafur/uracil; 5-FU: 5-fluorouracil: 5-DFUR: doxifluridine; HCFU: carmofur; ACT: adriamyacin, cyclophosphamide, taxol or taxotere; FECT: fluorouracil, epirubicin, cyclophosphamide, taxol; FECTH: FECT + hepirubicin; CT: cyclophosphamide, taxol; CAR: carmofur; MRI: magnetic resonance imaging; fMRI: functional MRI; PET: positron emission tomography; qEEG: quantitative electroencephalography.

Table 13. Pre-clinical studies focused on mechanism

Drug/Dose Mechanism Route/ frequency/duration Animals/cells Outcome Reference Doxorubicin (20mg/kg) Apoptosis, Single i.p. inj. (72 hours) Male B6C3 (2 mo.) Dox increased TNFa, pro-apoptotic proteins, and Tangpong Inflammation mice cell death; xanthone drivative abolished these 2011 effects Doxorubicin (0.01-1 Inflammation, 0-24 hours after treatment Human Urothelial cells Dox increased IL-8, IL-1b, PGE2, and basal ACH Kang mg/mL) Neurotransmitter (RT4) 2013 Doxorubicin (20mg/kg) Apoptosis, Single i.p. injection (3 Male B6C3 mice (2 Dox increased TNFa in brain, pro-apoptotic Tangpong Inflammation, hours) mo.) proteins, and induced mitochondrial dysfunction; 2006 Oxidative Stress inhibiting TNFa abolished these effects

Doxorubicin (20mg/kg) Inflammation, Single i.p. inj. (3 hour Male B6C3 and Dox increased TNFa in WT and iNOSKO and Tangpong Oxidative Stress treatment) iNOSKO decreased mitochondrial respiration in WT, but not 2007 iNOSKO mice. Dox also reduced MnSOD, a mitochondrial antioxidant enzyme Doxorubicin (45 mg/kg) Oxidative Stress Single i.v. (48 hours) Male Wistar rats Dox increased MDA, decreased GSH; melatonin Oz 2006 abolished these effects 8 Doxorubicin (20 mg/kg) Oxidative Stress Single i.p. inj. (72 hrs.) Male mice (2-3 mo.) Dox increases oxidative stress (protein oxidation, Joshi

lipid peroxidation, MRP1 expression) 2005 Doxorubicin (20 mg/kg) Oxidative Stress Single i.p. inj. (72 hrs.) Male B6C3 mice (2-3 Dox increases oxidative stress (protein oxidation, Joshi mo.) lipid peroxidation, MRP1 expression) and 2007 decreased glutathione (GSH). GCEE reversed these effects. Doxorubicin (20 mg/kg) Oxidative Stress ADR: i.p. inj. (72hr. before Male B6C3 mice (2-3 Dox decreased GSH and increased GPx, GST, and Joshi tissue collection) mo.) GR levels. Activity levels were increased for GPx 2010 and decreased for GST and GR. Dox increased oxidation of peroxiredoxen 1, TPI, and enolase proteins Doxorubicin (25 mg/kg) Oxidative Stress Single inj. Male B6C3 mice Dox increases oxidative stress and TNFa. 2- Aluise Mercaptoethane sulfonate prevents these effects. 2011 Cyclophosphamide (50 Neurogenesis/ 3 i.p. injections at day 1,4,7 C57BL/6 (2 mo.) All drugs reduced neural cell proliferation, but not Janelsins mg/kg), 5-FU (60 gliogenesis (tissue collected 48 hours apoptosis, in the dentate gyrus. Insulin-like growth 2010 mg/kg), dox (5 mg/kg), later) factor (IGF-1) slightly reduced this effect in mice or paclitaxel (5 mg/kg) treated with cyclo. Continued 80

Table 13 continued Drug/Dose Mechanism Route/ frequency/duration Animals/cells Outcome Reference Cyclophosphamide (30 Neurogenesis/ 1 i.v. doses every other day Male Lister-hooded rats Chemo reduced brdu (survival of hippocampal Lyons mg/kg) gliogenesis for 7 treatments cells), but not Ki67( proliferation); No difference in 2011 NLR behavior Cyclophosphamide (40 Neurogenesis Single i.p. inj. Male ICR mice (2 mo.) CP impairs learning/memory (NOR and passive Yang mg/kg) avoidance) and decreases neurogenesis for up to 2010 one week Cyclophosphamide Histone I.P. 1/week for 4 weeks Female Wistar rats (4 CMF decreased hipp. cell proliferation and histone Briones (40mg/kg) , 5-FU (75 acetylation, (behavior 2 weeks after last mo.) deacetylase activity, impaired learning/memory 2011 mg/kg), and MTX (37.5 Neurogenesis treatment) (MWM), and increased histone acetylation mg/kg) Cyclophosphamide Inflammation Four i.p. inj. 1/week (1 week Male athymic nude rats Both agents reduced neurogenesis and impaired Christie (50mg/kg) and before behavior) (2 mo.) learning/memory (NLR and fear conditioning). CP, 2012 doxorubicin (2 mg/kg) not DOX, activated microglia (ED-1+) Cyclophosphamide (75 Oxidative Stress Single i.p. inj. Male Swiss albino mice CP increases oxidative stress (decreases GSH, GSH Manda mg/kg) (6-8 weeks) peroxidase, and alkaline phosphtase). Melatonin 2003 reversed these effects. Cyclophosphamide Oxidative Stress Single i.p. inj. (24 hours) Wistar albino rats CP increased MDA in brain; reduced by hot short Oboh

8 (75mg/kg) pepper 2010

Cyclophosphamide Oxidative Stress Single i.p. inj. (24 hours) Wistar albino rats CP increased MDA, AST, ALT levels; reduced by Oboh (75mg/kg) yellow dye extract from Brimstone tree root 2012 Cyclophosphamide Oxidative Stress Single i.p. inj. (24 hours) Wistar albino rats CP increased lipid peroxidation (increased MDA Oboh (75mg/kg) levels); reduced by red dye extract from Sorghum 2010 stem Cyclophosphamide (75 Oxidative Stress Single i.p. inj. (24 hrs.) Male Swiss mice (2 CP increased oxidative stress; effects reduced by Bhatia mg/kg) mo.) Linseed oil. 2006 Continued

Table 13 continued Drug/Dose Mechanism Route/ frequency/duration Animals/cells Outcome Reference Cisplatin, Apoptosis, In-vitro: (24 hour treatment) In-vitro: neurons from In-vivo, all drugs induced neurotoxicity and cell Rzeski Cyclophosphamide, Morphology In-vivo: i.p. injection (4-24 19-day old male Wistar death. 2004 methotrexate, hours before tissue rats In-vitro, chemo drugs induced neurotoxicity, which vinblastin, thioTEPA collection) In-vivo: 7 day old were ameliorated by N-methyl-D-aspartate receptor In-vitro: 5-100,5-100,5- Wistar rats MK801, AMPAra GYKI52466, and pancaspase 100, 0.1-1,5-100 uM inhibitor Ac-DEVD-CHO In-vivo: cisplatin (5- 15mg/kg), cyclophosphamide (200-600), thiotepa (15- 45), ifosfamide (100- 500) Methotrexate (0.2 Oxidative Stress Daily subcutaneous inj. for 7 Male Wistar rats MTX increased oxidative stress (decreased GSH, Rajamani mg/kg) days (analysis performed 2 increased protein carbonyls, SOD, and catalase) 2006 days after last inj.)

Methotrexate (75 Neurogenesis/ 2 i.v. doses 1 week apart Male Lister-hooded rats MTX reduced hipp. cell survival, proliferation, and Lyons

2 mg/kg) gliogenesis NLR behavior; Fluoxetine prevented these effects 2011 Methotrexate (37.5- Neurogenesis/ Single i.v. inj. Male Wistar rats (3 MTX reduced cell proliferation and showed Seigers 300mg/kg) gliogenesis mo.) memory impairment in NLR and MWM probe 2008 latency, but not MWM training MTX (250 mg/kg) Neurogenesis/ Sing tail vein injection Male Wistar rats (3 MTX reduced hipp. cell proliferation 7 days after Seigers gliogenesis mo.) treatment, but not 1 day after; No diff. in cell death 2009 Methotrexate (100 Neurogenesis/ Single i.p. inj. Male buffalo rats (3 MTX reduces hippocampal cell proliferation Seigers mg/kg) gliogenesis mo.) whereas having a tumor did not 2010 Methotrexate (40 Neurogenesis/ Single i.p. inj. (24 hr. prior C3H/HeN tumor MTX increased depressive-like behavior (TST) and Yang mg/kg) gliogenesis, to behavior) bearing mice (6 weeks) cognitive impairment (passive avoidance), reduced 2012 oxidative stress hipp. neurogenesis, and upregulated iNOS and COX-2 enzymes MTX Apoptosis In-vitro: 24 hour treatment In-vitro: embryonic rat MTX-treated mice showed increased depressive- Yang In vitro: 0,40,or 200 uM Neurogenesis/ In-vivo: Single i.p. inj. (6 hippocampal cells like behavior (TST) and memory deficits (NOR) in 2011 In-vivo: 0-200 mg/kg gliogenesis hours to 14 days) In-vivo:Male 57BL/6 (2 addition to increased apoptosis and decreased cell mo.) proliferation in the hippocampus Continued

Table 13 continued

Drug/Dose Mechanism Route/ frequency/duration Animals/cells Outcome Reference Methotrexate + 5- FU Electrophysiolog I.P. 1/wk for 4 weeks Male C57BL/6Hsd EEG gating ratio was higher in both chemo groups Gandal High dose: 37, 75 y mice (7-8 wk old) at week 5; no difference ERP amplitude or latency 2008 mg/kg of P1 Low dose: 38, 19 mg/kg MTX (0.5mg/kg) CSF Composition Single intrathecal inj. in Male Long Evans rats MTX induced deficits in spatial and visual memory Vijayanat cistern magna (3 days before (8 weeks) (NOR and NLR) han 2011 behav.) OR 4 inj. over 10 MTX increased homocysteine sulfinic acid and days (1 month and 3 months) homocysteic acid in CSF, briefly in single inj. group and out to 3 months in multiple inj. group MTX (0.5mg/kg) CSF Composition Single intrathecal inj. in Male Wistar rats (3 MTX decreased folate and increased homocysteine Li 2010 cistern magna (3 days before mo.) in CSF. MTX induced deficits in spatial and visual Intrathecal behavior) memory (NOR and NLR) MTX CSF Composition Acute: single i.p. inj. (3 days Male Long Evan rats Acute and chronic MTX impaired spatial memory Li 2010 Acute: 250 mg/kg before behave.) (10 weeks) (NLR) but not visual memory (NOR) deficits and Systemic Chronic: 1mg/kg Chronic: 10 I.P. inj. over 8 reduced folate in serum and CSF. weeks (10 weeks after first inj.) MTX (250 mg/kg) Blood Supply Single i.v. inj. (1 or 3 weeks) Male Wistar rats MTX reduced hipp blood vessel density 1 and 3 Seigers

3 (3 mo.) weeks after treatment and metabolism (via PET) at 2010 1 but not 3 weeks; MTX reduced peripheral, but not hippocampal inflammatory cytokines, activated microglia at 1, 3 weeks. MTX (1-2 mg/kg) Neurotransmitter/ 1-5 i.c.v. inj. every 3 days Male Sprague-Dawley Single dose of MTX reduced metabolites of Silverstein monoamine for those receiving more than rats (4 weeks) dopamine, but not serotonin 6 hours after inj. 1986 release 1 inj. Multiple injections resulted in increased serotonin metabolites; no diff. on dopamine MTX (1.5-6 mg/kg) Neurotransmitter/ 3 daily injections in lateral Male Wistar rats MTX reduced brain amines (NorE, Dop, Ser), Madhyast monoamine ventricle (1 week before (4 mo. ) increased hipp. neuronal death, and induced ha 2002 release behavior) learning deficits (conditioned avoidance test), but did not affect anxiety-like behavior (light dark box) 5-FU (40 mg/kg) Apoptosis, In-vitro: 0.5-5uM for In-vitro: purified CNS In-vitro: 5-FU killed CNS cells but not cancer cells Han 2008 Inflammation, 24hours to 5 days cells from developing In-vivo: 5-FU increased apoptosis, decreased Electrophysiolog In-vivo: 3 i.p. inj. one every rat; cancer cell lines proliferation and induced delayed damage to y other day (1,7,14,56 days or In-vivo: CBA mice (6-8 myelinated tracts and auditory functioning; Neurogenesis/Gli 6 months) weeks) No difference in microglial activation (F4/80 ogenesis marker) 83 Continued

Table 13 continued

Drug/Dose Mechanism Route/ frequency/duration Animals/cells Outcome Reference 5-FU (25 mg/kg) Neurogenesis/ i.v. inj. every other day for 5 Male Lister hooded rats 5-FU impaired working memory (NLR and ElBeltagy gliogenesis total injections conditioned emotional response test(CER)) and 2010 reduced proliferation in dentate gyrus. Fluoxetine, an anti-depressant improved NLR and proliferation relative to 5-FU group 5-FU (25 mg/kg) Neurogenesis/ i.v. inj. every other day for 6 Male Lister hooded rats 5-FU reduced cell proliferation in dentate gyrus 2-4 ElBeltagy gliogenesis total injections weeks after, but not immediately following last 2010 treatment. COX-2 (inflammation) not corr. w/ cell proliferation or survival 5-FU (25 mg/kg) Neurogenesis/ 5 inj. over 2 weeks Male Lister-hooded rats 5-FU reduced hipp. cell survival, proliferation, and Lyons gliogenesis NLR behavior; Fluoxetine reversed these effects 2012 5-FU (20 mg/kg) Neurogenesis/ 5 i.v. inj. within 12 days Male Lister-hooded rats 5-FU reduced BDNF and doublecortin levels in the Mustafa gliogenesis hippocampus; no effect on spatial working memory 2008 (NLR) ThioTEPA (1,5, or 10 Neurogenesis/ I.P. inj. daily for 3 days C57BL/6 mice (6 wks.) thioTEPA, but not 5-FU, reduces cell proliferation Mignone mg/kg) OR gliogenesis old in the dentate gyrus of the hippocampus 2006

8 5-FU (50mg/kg)

ThioTEPA (10 mg/kg) Neurogenesis/ I.P. inj. daily for 3 days Male C57BL/6 (8 ThioTEPA reduced cell proliferation; learning Mondie gliogenesis weeks) (NOR) was impaired at 12 and 20 weeks, but not 2010 earlier; no effect in Porsolt or TST behavior ThioTEPA (10mg/kg) Neurogenesis/ Three daily i..p. inj. Male C57BL/6 (2 mo.) ThioTEPA reduced cell proliferation and increased Wilson gliogenesis depressive-like behavior in the sucrose anhedonia 2013 but not the Porsolt test. Cytosine arabinoside Morphology I.P. inj. for 5 consecutive Male Wistar rats (10 AraC impaired motor behavior (footprint pattern Koros (AraC; 400 mg/kg) days (1 day before behavior) weeks old) and rotarod) but did not affect learning (MWM). 2007 AraC also reduced cerebellar neurofillament activity. AraC (AraC; 400 Morphology I.P. inj. for 5 consecutive Male Wistar rats (10 AraC reduced neurofillament isoform H in the Koros mg/kg) days (tissue collection 24 weeks old) cerebellum. N-acetylcysteine (NAC), an 2009 hours after last inj.) antioxidant prevented most of this reduction. AraC (400 mg/kg) Morphology I.P. inj. for 5 consecutive Male Sprague-Dawley AraC treatment resulted in impaired long-term, but Li 2008 days (tissue collection 24 rats not short-term memory (MWM) and a reduced hours after last inj.) dendritic length, spine density and number of branch points in anterior cingulated cortex, but not hippocampal CA1. Continued 84

Table 13 continued

Drug/Dose Mechanism Route/ frequency/duration Animals/cells Outcome Reference Carmustine (BCNU; Neurogenesis/ Single treatment (1 hour Oligodendrocyte BCNU disrupted O-2A/OPC cell division and Hyrien 1uM) gliogenesis treatment) precursor cells (O- differentiation, increasing cell cycle length and 2010 2A/OPC) decreasing time to differentiation and likelihood of self-renewal Carmustine (BCNU), Apoptosis, In-vitro: BCNU 1 hour; In-vitro: neuroepithelial Chemo agents increased cell death, suppressed cell Dietrich cisplatin, cytarabine neurogenesis/ cisplatin 20 hrs.; cytarabine cells from rat spinal division 2006 In-vitro:5-200, .1-100, gliogenesis 24 hours cord; tumor cell lines In-vitro: chemotherapy agents are more toxic for .01-2 uM In-vivo: 3 consecutive i.p. In-vivo: CBA mice (6-8 CNS progenitor cells and nondividing In-vivo:10,5,250 mg/kg injections weeks) oligodendrocytes than cancer cell lines

BCNU (20mg/kg) Apoptosis, Single i.v. injection (3 weeks 60 male Wistar rats BCNU increases TNFa, apoptosis, oxidative stress, Helal Oxidative stress before behavior) cog. deficits; Metallothionein (i.c.v.) reduces these 2009 effects Cisplatin (0.1, .5, or 1 Electrophysiolog Daily i.p. inj. for 3 Male Sprague-Sawley High dose increased thermal and mechanical Cata 2008

8 mg/kg) y consecutive days rats withdrawal thresholds; Cisplatin (.5 mg/kg) 5 increased activity and frequency of spontaneous

neuronal firing in wide dynamic range neurons Cytosine-arabinoside Neurogenesis i.c.v. for 14 days via osmotic Male Sprague-Dawley Cytosine-arabinoside suppresses neurogenesis and Lau 2009 (2%) pump rats impaired learning/memory (fear conditioning) Vincristine Morphology Single i.c.v. injection in the Male Wistar rats (2 Chemo had a dose dependant “all or none” effect Eijkenboo 1st study: 0.5-2uL of dorsal hippocampus mo.) on learning (MWM). Low dose had no effect, high m 1999 0.55mg/mL dose group didn’t learn the task at all. Not a good 2nd study: 1uL of 0.06, drug for a learning/memory model 0.18, or 0.55 mg/mL Temozolomide (TMZ; Electrophysiolog I.P. inj. 3 consecutive days Male Sprague-Dawley TMZ reduced endogenous theta wave activity Nokia 25 mg/kg ) y each week for 6 weeks rats (2 mo.) (associated with learning/memory), reduced 2012 neurogenesis, and impaired learning and memory (classical eye-blink task) BCNU: Carmustine; GSH: Glutathione; GCEE: y-glutamyl cysteine ethyl ester; MDA: malondialdehyde; SOD: superoxide dismutase; MRP1: multidrug resistance- associated protein; TPI: triose phosphate isomerase; AST:aspartate aminotransferase; ALT: alanini amino-transferase; SOD: superoxide dismutase; PGE2: prostaglandin E2 ; ACH: acetylcholine.

Chapter 2: Methods

All procedures were approved by the Ohio State University and were conducted in accordance with National Institutes of Health guidelines for the

Care and Use of Laboratory Animals (8th Edition) of animals and under protocols approved by the institutional animal care and use committee.

Animals and environment: Adult female BalbC (Charles River, Wilmington, MA) or C57BL6 (Charles River, Wilmington, MA) mice were housed in a temperature and humidity-controlled, ALAAC approved vivarium on a 14:10 light/dark cycle.

All animals were allowed ad libitum access to food and water. All females were ovariectomized at least two weeks prior to being used in an experiment.

Minocycline administration: Minocycline was dissolved in the drinking water.

The concentration of minocycline was adjusted on a daily basis to achieve a dose of approximately 90mg/kg/day. Mice received minocycline or the control water beginning two days prior to chemotherapy treatment, continuing through the end of the study. The minocycline and control water were prepared fresh daily and the total liquid consumed was recorded.

Chemotherapy administration: Animals received a single intravenous tail injection of either a combination of Doxorubicin (13.5mg/kg; TEVA Parenteral

Medicines) and Cyclophosphamide (135 mg/kg; Baxter) or an equivalent volume of the vehicle (0.9% isotonic saline).

Intracerebroventricular Cannulation: A stereotaxic apparatus was used to implant a cannula into the left lateral ventricle of anesthetized mice (isoflurane) 2 days before chemotherapy treatment (cannula position: +0.02 posterior and –

0.95 lateral to bregma, extending 2.75 mm below the skull; Plastics One,

Roanoke, Va). The cannula was connected by tubing to an Alzet minipump

(Model 1002; Durect, Cupertino, Calif) that was implanted in the left lateral ventricle for constant infusion (0.25µL/day) of vehicle (artificial cerebrospinal fluid

[aCSF]) or minocycline (10ug/day; <2µl) for 6 days. Tissue was then collected and analyzed for proper cannulation placement.

Behavioral Tests:

Locomotor Activity: Overall locomotor activity was tested using the open field, a 40 x 40 cm clear acrylic chamber lined with clean corncob bedding, inside a ventilated cabinet (Med Associates, St Albans, VT, USA). A frame at the base of the chamber consisting of 32 photobeams in a 16 x 16 arrangement, in addition to a row of beams above, detected the location of horizontal movements and rearing, respectively (Open Field Photobeam Activity System, San Diego

Instruments, San Diego, CA, USA). Total movement was tracked for 30 min and analyzed for the following: (1) the percentage of beam breaks in the center

(inside 30 x 30 cm) of the open field, (2) number of rears and (3) total locomotor activity.

Elevated Plus Maze: Anxiety-like behavior was assessed in the elevated plus maze, which is a plus-shaped apparatus, elevated above the ﬂoor with two dark enclosed arms and two open arms. Each mouse was placed in the center of the test apparatus to begin and behavior was recorded on video for 5 min. An uninformed observer later scored videotapes for the number of entries and time spent in both the open and closed arms of the maze using Observer software.

Forced Swim test: To assess depressive-like behavioral responses in the forced swim test (Porsolt, Le Pichon & Jalfre, 1977), mice were placed in a clear 4 liter container ﬁlled with room-temperature water (22 ± 1ºC) for 3 min. Behavior was recorded on video and subsequently scored with Observer software (Noldus,

Wageningen, Netherlands) by an observer unaware of assignment to experimental groups. The behaviors scored were: (1) swimming or (2)

ﬂoating/immobility (i.e., minimal movement necessary to keep head elevated above water surface).

Learning and memory: Learning and memory was assessed in the Barnes maze, a white circular polypropylene disk 91 cm in diameter (San Diego

Instruments, CA), elevated 1.2 meters above the floor. The disk has 18 holes, 5 cm in diameter, evenly spaced around the perimeter. One of the holes leads to a dark escape box. Noldus tracking software (Leesburg, VA) records latency to reach the escape hole, path length, and number of errors. The testing occurs in a brightly lit room between one and six hours after the onset of the light phase in the colony. Mice received six days of training, consisting of three trials per day; the intertribal interval was approximately 10 min. If the mouse did not find the hole within the first two minutes of a trial, the latency was recorded as 120 seconds and the mouse was guided to the escape hole. A probe trial (90 s) was conducted on Day 7, in which the escape box was removed. Percent of visits to the target hole and latency to first reach the target hole were recorded. Beginning

3 hours after the probe trial, mice received 3 days of reversal trials in which the escape box was rotated 180 degrees away from its location; the reversal training also consisted of three trials per day.

Microglia enrichment: Brain tissue was minced and homogenized as previously described (Henry, Huang, Wynne, & Godbout, 2009). Single cell suspensions were resuspended in 70% Pecoll to which 50% Percoll, 35% Percoll, and 0%

Percoll were layered atop respectively to generate a density gradient during a 20

89 minute centrifugation. The interface between the 70% and 50% layers consisting of enriched microglia was removed, washed and used for subsequent studies.

LPS Microglial Stimulation: Enriched microglia were resuspended in BV2 media, plated in a poly-l-lysine coated 96 well plate at 100,000 cells in 200uL media per well and placed in an incubator (37ºC w/ 5% CO2) for one hour.

Following incubation each well received 10uL of saline (control) or 10uL 0.40

µg/ml LPS (Sigma, St. Louis, MO) dissolved in sterile saline for 18 h at 37°C and

5% CO2. Supernatants were collected and stored at -80 C for multiplex cytokine analysis and microglial cells were collected for PCR analysis.

Protein Extraction and Determination of Cytokine Concentrations: Brains were removed and processed through a 70um filter to generate a single cell suspension. After the cells were pelleted by centrifugation, cells were disrupted in

300uL of RIPA buffer (containing 0.5% bovine serum albumin) followed by sonication for five 3 second pulses. The homogenate was again spun down and the supernatant was collected, aliquoted and stored at -80°C until assay. The protein concentration of each sample was measured using the Pierce BCA protein assay (Thermo Scientific, Rockford, IL). The concentrations of IL-1ß, IL-2,

IL-4, IL-5, IL-6, IL-10, interferon γ (IFN-γ), tumor necrosis factor α (TNF-α and granulocyte macrophage colony-stimulating factor (GM-CSF) were measured using the Mouse Cytokine Group 1 kit (Biorad kit # M60-000011J) according to

90 the manufacturer’s protocol. For cytokine concentrations in the blood, IL-1α, IL-

1ß, IL-2, IL-3, IL-4, IL-5, IL-6, IL-9, IL-10, IL-12(P40), IL-12(P70),IL-13, IL-17, eotaxin, granulocyte colony-stimulating factor (G-CSF), KC, MCP-1, MIPα, MIP-

1ß, RANTES, IFN-γ, TNF-α, and GM-CSF were measured using the Mouse

Cytokine Group 1 Panel 23-plex kit (Biorad kit # M60-009RDPD). All samples, standards, and blanks were run in duplicate using a Bio-plex 200 system plate reader and Bio-plex manager 6.1 software (Hercules, CA).

Real-time PCR for tissue: Brains were removed and stored in RNA later for four days at 0 degrees Celsius. The hippocampus and prefrontal cortex were removed and total RNA was extracted from these regions using a homogenizer

(Ultra-Turrax T8, IKA Works, Wilmington, NC) and an RNeasy Mini Kit (Qiagen,

Valencia CA) according to manufacturer’s protocol. Extracted RNA was suspended in 30µL of RNase-free water and RNA concentration was determined by a spectrophotometer (NanoDrop ND-1000, Wilmington, DE). A TaqMan 18S rRNA primer and probe set (labeled with VIC dye: Applied Biosystems, Foster

City, CA) were used as a control gene for relative quantification in addition to the primers and probes for the desired genes of interest (Applied Biosystems, Foster

City, CA). Amplification was performed on an ABI 7000 Sequencing Detection

System by using Taqman Universal PCR master mix. The universal two-step

RT-PCR cycling conditions used were: 50°C for 2min, 95°C for 10min followed by

40 cycles of 95°C for 15sec and 60°C for 1min.

Real-Time PCR for cells: Total RNA was extracted from enriched microglia by using a sonicater and an RNeasy Micro Kit (Qiagen, Valencia, CA) according to manufacturer’s protocol. Extracted RNA was suspended in 15 μL of RNase-free water, and RNA concentration was determined by a spectrophotometer

(NanoDrop ND-1000). cDNA was made and PCR run as described above.

Flow Cytometry for brain and spleen: Brain tissue was obtained immediately following euthanasia and passed through 70uM cell strainers to obtain single-cell suspensions. Red blood cells were lysed and remaining cells were washed and resuspended in stain wash buffer. Cell surface Fc receptors were blocked via incubation with anti-CD16/32 antibody (ebioscience, San Diego, CA) and then washed with stain wash buffer. All antibody incubations were performed on ice in the absence of light. In general, the cells were incubated with mixtures of specific concentrations of fluorochrome-labeled antibodies according to experimental design. Cells were fixed in 2% paraformaldehyde in PBS and flow cytometry data was acquired using a BD LSRII instrument and analyzed using

TreeStar FlowJo software (Ashland, Or). For any given marker, all of the analysis gates were identical in size and position.

Flow Cytometry for microglia: Microglia were enriched as previously described and placed in 5 mL tubes followed by Fc receptor blocking and staining as

92 described above. Flow cytometry data were acquired using a BD LSRII instrument (San Jose, CA) and analyzed using TreeStar FlowJo software

(Ashland, Or). For any given marker, all of the analysis gates were identical in size and position.

Hippocampal morphological analyses: Mice were euthanized between 2 and 6 hours after the onset of the light phase in the colony room; brains were removed and impregnated using the FD Rapid GolgiStain Kit (FD NeuroTechnologies,

Ellicott City, MD, USA) according to the manufacturer’s instructions. Brains were sliced at 100 mm, thaw mounted onto gelatin-coated slides that were coded in order to remove observer bias, counterstained with cresyl violet (Sigma, St Louis,

MO, USA), dehydrated and coverslipped. Brains were assessed for hippocampal cell morphology and spine density in three hippocampal subfields: dentate gyrus,

CA1 and CA3. Sections were visualized using a Nikon E800 brightfield microscope (Nikon, Burlington, VT, USA) and traced using Neurolucida software

(MicroBrightField, Burlington, VT, USA) at a magnification of 200x for neuronal morphology and 1000 x for spine density. Four to five representative neurons were selected per area, per animal for tracing. Selected neurons were fully impregnated and lacked truncated dendrites. Whole cell traces were analyzed using NeuroExplorer software for cell body size and perimeter and dendritic length (MicroBrightField). Sholl analyses were conducted separately for apical and basilar dendrites. From each of the five neurons selected per animal four

93 segments > 20 mm were selected in the apical and basilar areas, respectively,

(except in the dentate gyrus where granule cells lack bidirectional projections).

All spine segments selected were at least 50 mm distal to the cell body. Spine density (spines per 1 mm) was calculated for each trace and averaged per cell, per area and per animal.

Immunohistochemistry: Mice were anesthetized (pentobarbital) and perfused transcardially with 0.1 mol/L phosphate-buffered saline followed by 4% paraformaldehyde. Brains were removed, post-fixed for 4 h, and then cryoprotected in 30% sucrose in 0.2 mol/L phosphatebuffered saline. The brains were sectioned coronally at 40 μm on a freezing microtome and rinsed 3 times for 5 minutes each in 0.1 mol/L phosphate-buffered saline. The sections were then incubated for 40 hours at 4°C with the appropriate antibodies. After 3 five minute rinses, the tissue was incubated with secondary antibodies for 2 days at room temperature. The tissue was again rinsed 3 times and then mounted onto

Super Frost Plus slides (Fisher, Hampton, NH, USA). All images were captured using a Zeiss LSM 510 confocal microscope.

Chemotherapy concentration in brain and blood: Liquid chromatography tandem mass spectrometry (LC-MS/MS) was used to determine the concentration of doxorubicin and cyclophosphamide agents in the brain and blood from five minutes to twenty-four hours after treatment. After blood and

94 brain were collected, 200μL plasma sample was prepared immediately from blood. A 100uL plasma aliquot for cytoxan/ phosphoramide mustard assay was treated with semicarbazide to prevent 4-hydroxy-cytoxan from converting to phosphoramide mustard, and the other 100uL plasma aliquot was transferred into a separated micro centrifuge tube for assay of doxorubicin. Brain extract sample for cytoxan/ phosphoramide mustard assay was prepared from 100mg brain tissue homogenized with 250μL PBS and 25μL 2M semicarbazide solution; for doxorubicin the same amount of brain tissue was homogenized in 250μL of

50% MeOH/ 50% H2O. All the procedures were carried out on ice, all solutions pre-chilled to 4°C, and all samples were snap frozen on dry ice and stored at -

80 C.

Two bioanalytical methods were developed and validated for quantification of doxorubicin, cytoxan and its active metabolite phosphoramide mustard in mouse plasma and brain using liquid chromatography tandem mass spectrometry (LC-MS/MS), one for simultaneous determination of doxorubicin, the other for cytoxan and phosphoramide mustard. Briefly, the cytoxan and phosphoramide mustard were extracted from plasma and brain extract samples with solid phase extraction (SPE) using Waters Oasis MAX cartridges.

Doxorubicin was recovered from plasma and brain extract samples by protein precipitation with methanol. A Thermo Dionex RSLC system consisted of degasser, ternary pump, column oven and autosampler was used for solvent and sample delivery. A TSQ Quantum Ultra EMR mass spectrometer equipped with

95 an electrospray ionization (ESI) source was used for mass analysis and detection. Both methods used an Agilent ZORBAX Extended C18 column

(50×2.1mm I.D, 3.5μm particle size) coupled with an Agilent Extend C18 guard cartridge (2.1mm×12.5mm, 5μm particle size) at 40 C with a total run time of

6.0min. The chromatographic separation was performed using two linear gradient elution programs with mobile phases consisted of H2O with 0.2% formic acid and methanol with 0.2% formic acid at a flow rate of 300μL/min for both assays. The two methods carried out the MS detection with selected reaction monitoring under positive mode using transitions from the precursor ions to product ions at m/z 544.118-397.02 for doxorubicin, m/z 261.02-140.07 for cytoxan and m/z

221.02>142.05 for phosphoramide mustard. These assays were developed and conducted by the Pharmacoanalytic Shared Resource of the Comprehensive

Cancer Center, OSU.

Data analysis: Student’s t-tests were used to analyze OF, EPM, FST, PCR, flow cytometry, neuronal morphology, and bioplex when comparing between two experimental groups (chemotherapy versus vehicle). A one-way analysis of variance (ANOVA) was used to analyze OF, EPM, FST, PCR, flow cytometry, bioplex, and neuronal morphology when analyzing the effects of minocycline on chemotherapy; LSD post-hoc was used to assess between group differences.

For Barnes maze analysis repeated-measures ANOVA were conducted with drinking and drug as the with-in subject factors and session as the between

96 subject factor to analyze latency to find the target hole, error rate, and path length. Following a significant result on repeated measures ANOVAs, comparisons were made between groups using one-tailed LSD post-hoc tests based on a priori hypotheses. Differences were considered statistically significant if p values were less than 0.05. All error bars represent standard error of the mean (SEM). Statistical analyses were conducted with SPSS software (version

22).

Chapter 3: Chemotherapy and Affective Behavior

While findings from clinical studies clearly suggest that cancer is associated with increased anxiety and depression, relatively few studies have assessed the effects of chemotherapy on depression and anxiety. Only recently has it been suggested that BC patients treated with chemotherapy are at increased risk of depression and anxiety relative to BC patients who are treated with surgery, radiotherapy or endocrine therapy (Redeker, Lev, & Ruggiero,

2000; Reece, Chan, Herbert, Gralow, & Fann, 2013; Torres et al., 2013). Up to forty percent of BC patients treated with chemotherapy suffer from depression and 27% from anxiety (Hack et al.; So et al., 2010). Furthermore, some specific drug regimens may be associated with increased risk; for example, treatment with a doxorubicin based regimen was nearly three times as likely to result in

97 depressive symptoms as a chemotherapy regimen not containing doxorubicin or a non-chemotherapy treatment (Avis et al., 2012). Similarly, a longitudinal study reported an increase in depression when treated with doxorubicin/cyclophosphamide, but not other regimens (Reece, Chan, Herbert,

Gralow, & Fann, 2013). These data are particularly concerning because doxorubicin/ cyclophosphamide (AC) is a commonly used and effective regimen for treating BC. In addition to the obvious concern regarding the negative effect of affective disorders on the quality of life for BC patients, depression also has been associated with fatigue, sleep disturbance, decreased survival, and increased tumor growth and recurrence rates (Falagas et al., 2007; Goodwin,

Zhang, & Ostir, 2004; Khan, Amatya, Pallant, & Rajapaksa, 2012; Liu et al.,

2009; Meyer, Sinnott, & Seed, 2003; Redeker, Lev, & Ruggiero, 2000).

Furthermore, depression can interfere with medical compliance (Colleoni et al.,

2000). An additional impediment to treating depression in BC survivors is that several commonly prescribed classes of antidepressants can interfere with the efficacy of ET (Kelly et al., 2010). Thus, treating depression and anxiety in BC patients is currently a challenge.

It is not known why doxorubicin/cyclophosphamide is more commonly associated with chemotherapy-associated depression than other regimens, although it may be related to the inflammatory properties of doxorubicin

(Tangpong et al., 2006). Findings from pre-clinical studies suggest that doxorubicin increases TNFα, IL-1b, and IL-8 in rodent models (Aluise et al.,

2011; Kang, Chess-Williams, Anoopkumar-Dukie, & McDermott, 2013; Niiya et al., 2003; Tangpong et al., 2011) and in vitro using human lung carcinoma cells

(Niiya et al., 2003). Furthermore, in a clinical study BC patients treated with doxorubicin, cyclophosphamide, and 5-fouorouracil had higher concentrations of

IL-6 than patients treated with cyclophosphamide, methotrexate, and 5- fluorouracil (Janelsins et al., 2012).

Pro-inflammatory cytokines are associated with depression in otherwise healthy individuals, as those with major depressive disorder have increased serum concentrations of IL-1, IL-6, and TNF-α (Maes et al., 1997; Mikova,

Yakimova, Bosmans, Kenis, & Maes, 2001; Miller, Maletic, & Raison, 2009).

More specifically, BC studies also report an association between depression and increased serum concentrations of IL-6 and inflammatory mediator NF-ĸB (Jehn et al.; Musselman et al., 2001; Torres et al., 2013). However, as all of these studies are correlative, they do not assess causation. Numerous pre-clinical studies, using a wide range of models, report that upregulation of the immune system can induce sickness behavior and depressive-like behavior (Dantzer,

O'Connor, Freund, Johnson, & Kelley, 2008; Neigh et al., 2009; Norman et al.,

2010; O'Connor et al., 2009; Wynne, Henry, & Godbout, 2009). Furthermore, inhibiting the increase in inflammation also ameliorates the depressive-like and anxiety-like behavior (Norman et al., 2010). For example, peripheral nerve injury- induced inflammation (IL-1b) and depressive-like behavior after spared nerve injury are ameliorated with administration of IL-1 receptor antagonist (Norman et

99 al., 2010). In a cardiac arrest mouse model, minocycline reduced ischemia- related microglial activation and anxiety-like behavior (Neigh et al., 2009). Social defeat also increases IL-6 and TNF-α expression and induces anxiety-like behavior; however, social defeat appears to be mediated by IL-1 since social defeat does not increase microglial activation or anxiety-like behavior in IL-1 receptor type-1-deficient mice (Wohleb et al.). Administration of lipopolysaccharide (LPS) has also been used to activate the immune system and induces depressive-like behavior that is ameliorated by minocycline (O'Connor et al., 2009). Thus, these data provide strong evidence for a causal relationship between elevated proinflammatory cytokines in the brain and affective changes in rodents. Taken together with the clinical studies that demonstrate correlative associations between serum cytokines and depression in BC patients being treated with chemotherapy (Torres et al., 2013) and the proinflammatory nature of AC-based chemotherapy (Janelsins et al., 2012; Tangpong et al., 2011), it is reasonable to hypothesize that chemotherapy-induced neuroinflammation underlies the development of depressive-like and anxiety-like behavior.

The goals of this study are 1) to determine if the affective consequences of chemotherapy can be recapitulated in mice, 2) to determine if chemotherapy produces neuroinflammation, and 3) to establish whether a causal relationship exists between chemotherapy-induced neuroinflammation and changes in affective behavior. If inflammation is part of the causal mechanism linking chemotherapy and depressive-like behavior, then this research could offer new

100 therapeutic targets for the treatment of depression in BC patients. Alleviating depression could not only improve the quality of life for breast cancer patients, but also potentially lead to better health outcomes.

Experimental Design

Experiment 1. Ten adult Balb/C female mice were ovariectomized two weeks before the experiment. Seven mice received a single tail vein injection of doxorubicin (13.5 mg/kg) and cyclophosphamide (135 mg/kg) chemotherapeutic agents (AC); five received saline (vehicle). Three days after treatment, brains were collected; a small section of prefrontal cortex was removed from alternating hemispheres for PCR analysis. The remaining tissue was homogenized and prepared for analysis of cytokines via a bioplex assay.

Experiment 2. Fifteen Balb/C female mice were ovariectomized two weeks before the experiment. Three days before chemotherapy treatment, females were separated and housed one mouse per cage for the remainder of the experiment. Eight mice received a single tail vein injection of AC; seven received vehicle. Three days later, following intravenous injection, behavioral testing was performed between one and six hours after the onset of the light phase in the colony room; mice were allowed to habituate to the test room for 30 min prior to testing. Tests were performed in the following order: (1) open field (2) elevated plus maze, (3) forced swim test; the order of testing reflects arrangement from

101 the least potentially intrusive to most intrusive test. The mice were returned to their cage for ten minutes between tests.

Experiment 3. Twelve C57BL/6 adult female mice were ovariectomized two weeks before the experiment. Seven mice received a single tail vein injection of

AC; five received the vehicle. Three days after treatment, behavior was assessed between 06:00 and 10:00 using the open field, elevated plus maze, and forced swim task, in that order. After each test mice were placed back in their cage for ten minutes before beginning the next test. One hour after behavior, brains were collected and prepared for analysis using flow cytometry.

Experiment 4. Forty Balb/C two month old female mice were ovariectomized two weeks before the experiment. Mice were injected with AC in the tail vein.

Brains and blood were collected at various time points after treatment (5 min., 10 min., 15 min., 30 min., 1, 1.5, 2, 4, 6, 8, or 24 hours) with four mice per group.

Serum and brain were processed for LC-MS/MS to determine drug concentration in the blood and brain.

Experiment 5.

Twenty Balb/C adult female mice were ovariectomized two weeks before the experiment. Two days before receiving AC treatment or vehicle (tail vein), mice were implanted with a cannula in the right lateral ventricle, receiving aCSF or

102 minocycline (10ug/day at .25uL/hour) throughout the remainder of the experiment. Three days following AC treatment, behavior was assessed between

06:00 and 10:00 using the open field, elevated plus maze, and forced swim task, in that order. After each test mice were placed back in their cage for ten minutes before beginning the next test. One hour after behavior, brains were collected and prepared for PCR analysis of mRNA.

Experiment 6.

Forty Balb/C two month old female mice were ovariectomized two weeks before the experiment. Beginning two days before receiving AC treatment (tail vein) and remaining until the end of the study, mice were administered either minocycline

(90 mg/kg) in their drinking water or the typical water; the water was changed every day for each animal. Three days following AC treatment, behavior was assessed between 06:00 and 10:00 using the open field, elevated plus maze, and forced swim task, in that order. After each test mice were placed back in their cage for ten minutes before beginning the next test. One hour after behavior, brains were collected and prepared for PCR analysis of mRNA.

103

Results

Experiment 1

Cytokine Concentration in the brain

Chemotherapy treatment upregulated protein expression of several pro- inflammatory cytokines within the brain (Figure 3.1). Three days after injections, chemotherapy treated mice had higher concentrations of pro-inflammatory cytokines tumor-necrosis alpha (TNFα; t8=3.5, p<.01), and chemokine (C-C motif) ligand four (CCL4; t8=2.2, p<.05) in the brain relative to control mice. IL-

1ß, IL-2, IL-4, IL-5, IL-6, IL-10, interferon γ (IFN-γ), and granulocyte macrophage colony-stimulating factor (GM-CSF), other pro-inflammatory markers, as well as anti-inflammatory marker IL-10 were unaffected by chemotherapy (p>0.05).

Pro-inflammatory Gene Expression in the Prefrontal Cortex of the Brain

Chemotherapy upregulated gene expression of several pro-inflammatory cytokines (Figure 3.2). Relative gene expression of TNFα (Figure 3.5 A; t6=2.1, p<.05) and IL-6 (Figure 3.5B; t8=4.1, p<.01) were increased in chemotherapy treated mice as compared to those injected with the vehicle.

Experiment 2

Depressive-like Behavior

Chemotherapy increased the time engaged in immobility, or behavioral despair, in the forced swim test (Figure 3.3A). Specifically, relative to vehicle

104 treated controls, mice treated with a combination of doxorubicin and cyclophosphamide spent significantly more time floating than actively swimming

(t13=2.4, p<.05), indicative of increased behavioral despair. Chemotherapy did not have an effect on the number of floating bouts (Figure 3.3B; t13=0.9, p>.05) or latency to the first float (Figure 3.3C; t13=1.4, p>.05), suggesting that chemotherapy-induced lethargy was not a factor.

Anxiety-like Behavior

Mice treated with chemotherapy showed anxiogenic behavior (Figure 3.4

A) as they spent less time exploring the open arms of the elevated plus maze relative to controls (t13=2.7, p<.05). However, there was no chemotherapy effect on locomotor activity (Figure 3.4 B) as there was no difference between groups in the total number of arms entered (closed and open combined; t13=1.1, p>.05).

Locomotor Activity

Chemotherapy does not affect overall locomotor activity (Figure 3.5). Mice treated with chemotherapy showed no difference in the total number of beams crossed in the open field chamber, indicative of locomotor activity, relative to vehicle treated mice (t13=1.9, p>.05).

105

Experiment 3

Depressive-like Behavior in C57bl/6 Mice

Chemotherapy increased the time engaged in immobility, or behavioral despair, in the forced swim test (Figure 3.6A). Specifically, relative to vehicle treated controls, mice treated with a combination of doxorubicin and cyclophosphamide spent significantly more time floating than actively swimming

(t10=2.1, p<.05), indicative of increased behavioral despair. Chemotherapy did not have an effect on the number of floating bouts (Figure 3.6AB; t10=0.41, p>.05) or latency to the first float (Figure 3.6C; t10=0.7, p>.05), suggesting that chemotherapy-induced lethargy was not involved.

Anxiety-like Behavior in C57bl/6 Mice

Mice treated with chemotherapy showed anxiogenic behavior (Figure 3.7

A) as they spent less time exploring the open arms of the elevated plus maze relative to controls (t10=1.9, p<.05). However, there was no chemotherapy effect on locomotor activity (Figure 3.7 B) as there was no difference between groups in the total number of arms entered (closed and open combined; t10=1.1, p>.05).

Locomotor Activity in C57bl/6 Mice

Chemotherapy does not affect overall locomotor activity (Figure 3.8). Mice treated with chemotherapy showed no difference in the total number of beams

106 crossed in the open field chamber, indicative of locomotor activity, relative to vehicle treated mice (t10=1.2, p>.05).

Protein Expression of Microglial Activation Markers

Chemotherapy increased protein expression of microglial activation markers within the brain (Figure 3.9). Three days following a single treatment of chemotherapy C57bl/6 mice had more cd11b+ cells (t10=2.5, p<.05) and increased protein expression of microglial activation markers CD80 (t8=3.1, p<.05) and CD86 (t8=2.2, p<.05) compared to control mice. There was no difference in expression of MHCII (t9=0.9, p>.05).

Experiment 4

Time course concentrations of doxorubicin and cyclophosphamide

Cyclophosphamide and its metabolite, phosphamide, are present in both the blood (Figure 3.10 A) and brain (Figure 3.10 B) within five minutes of i.v. administration. However, two hours after injection, only trace amounts are present in the brain, none in the blood. Doxorubicin is present in the blood up to

24 hours after injection, but not detected in the brain at any time point.

107

Experiment 5

Depressive-like behavior after i.c.v. minocycline administration

Minocycline ameliorated chemotherapy-induced immobility, or behavioral despair, in the forced swim test (Figure 3.11A; F3,15=7.3, p<.01). Post-hoc analysis revealed that mice receiving minocycline and chemotherapy spent less time floating than chemo treated mice receiving aCSF (vehicle), suggesting that minocycline ameliorates the depressive-like behavior induced by chemotherapy.

The number of floating bouts differed (Figure 3.11B; F3,15 =6.7, p<.05) in that both chemo groups floated more times than the vehicle groups (P<.05). There was no difference between groups in latency to first float (Figure 3.11C; F3,15

=1.1, p>.05).

Anxiety-like behavior after i.c.v. minocycline administration

Minocycline ameliorated chemotherapy-induced suppression of anxiogenic-like behavior, in the elevated plus maze (Figure 3.12; F3,15=10, p<.01). Post-hoc analysis revealed that mice receiving minocycline and chemotherapy spent more time in the open arms than chemo treated mice receiving aCSF (vehicle), suggesting that minocycline ameliorates the anxiety- like behavior induced by chemotherapy.

108

Locomotor Activity

Minocycline (i.c.v.) does not affect overall locomotor activity (Figure 3.13).

There were no differences in locomotor activity in mice receiving chemotherapy and/or minocycline (Figure 3.14; F3,14=1.1, p>.05).

Effects of minocycline on gene expression of pro-inflammatory markers after AC

Minocycline (i.c.v.) ameliorated chemotherapy-induced gene expression of pro- inflammatory cytokine IL-6 in the pre-frontal cortex (Figure 3.14; F3,15=9, p<.01).

Post-hoc analysis revealed that mice receiving minocycline and chemotherapy had reduced gene expression of pro-inflammatory marker IL-6 relative to mice receiving normal drinking water and chemotherapy (p < .05). There were no differences for IL-1b or TNF-α (data not shown).

Experiment 6

Depressive-like behavior after oral minocycline administration

Minocycline (oral administration) ameliorated chemotherapy-induced immobility, or behavioral despair, in the forced swim test (Figure 3.15A; F3,35=4.2, p<.05). Post-hoc analysis revealed that mice receiving minocycline and chemotherapy spent less time floating than chemotherapy treated mice receiving aCSF (vehicle), suggesting that minocycline ameliorates the depressive-like behavior induced by chemotherapy. The number of floating bouts differed (Figure

109

3.15B; F3,35 =4.1, p<.05) in that the chemotherapy water group had more floating bouts than the vehicle water and chemotherapy minocycline (p<.05), but not vehicle minocycline (P>.05) groups. The vehicle minocycline group also had more floating bouts than the chemotherapy minocycline (p<.05), but not vehicle water (p>.05) groups. Given there is no specific effect of chemotherapy or minocycline on number of floating bouts and the chemotherapy minocycline group performed better than both the chemo water and vehicle minocycline groups, these results suggest that the number of floating bouts may not be a good measure of depressive-like behavior in this model. There was no difference between groups in latency to first float (Figure 3.15C; F3,15 =1.1, p>.05).

Locomotor Activity

Minocycline (oral administration) does not affect overall locomotor activity

(Figure 3.16). There were no differences in locomotor activity in mice receiving chemotherapy and/or minocycline (Figure 3.14; F3,14=1.1, p>.05).

Effects of minocycline on gene expression of pro-inflammatory markers after AC

Minocycline (oral administration) ameliorated chemotherapy-induced gene expression of pro-inflammatory cytokine IL-6 in the pre-frontal cortex (Figure

3.17; F3,15=9, p<.01). Post-hoc analysis revealed that mice receiving minocycline and chemotherapy had reduced gene expression of pro-inflammatory marker IL-

110

6 relative to mice receiving normal drinking water and chemotherapy (p < .05).

There were no differences for IL-1b or TNF-α (data not shown).

Discussion

Clinical studies suggest that treatment with doxorubicin and cyclophosphamide are more likely to result in depression and anxiety than a regimen using other chemotherapeutic agents (Avis et al., 2012; Reece, Chan, Herbert, Gralow, &

Fann, 2013). Thus, I hypothesized that non-tumor bearing mice treated with doxorubicin and cyclophosphamide would also show an increase in depressive- like and anxiety-like behavior relative to vehicle treated mice. Indeed, relative to control mice, chemotherapy treated mice spent more time floating in the forced swim test, indicative of depressive-like behavior. Additionally, chemotherapy treated mice spent less time in the open arms of the elevated plus maze, indicative of anxiety-like behavior. Both BalbC and C57bl/6 mice were tested in all three tasks and showed similar results, suggesting that chemotherapy affects both of these strains in a similar manner. Importantly, results from the open field test suggest that chemotherapy did not have an impact on overall locomotor activity, thus it is unlikely that chemotherapy-induced lethargy was a confounding factor in these tests.

As a few clinical studies have found that chemotherapy is associated with an increase in serum proinflammatory markers (S. Kesler et al., 2013; Torres et al., 2013), we hypothesized that treatment with concomitant doxorubicin and

111 cyclophosphamide would increase proinflammatory markers within the brain.

Indeed, three days after a single dose of chemotherapy gene expression, pro- inflammatory cytokines TNF-α and IL-6 and protein concentrations of TNF- α and

CCL4 were higher relative to vehicle treated mice. Changes in affective behavior have been shown to be induced by an increase in pro-inflammatory cytokines secreted by activated microglia (Dantzer, O'Connor, Freund, Johnson, & Kelley,

2008; Wynne, Henry, & Godbout, 2009). Thus, we hypothesized that chemotherapy treatment would result in an increase in microglial activation markers. Indeed, microglia from mice treated with doxorubicin and cyclophosphamide expressed higher levels of activation markers CD80 and

CD86.

CCL-4, a chemokine mainly known to be secreted by macrophages in the periphery, is also expressed by microglia and astrocytes and is associated with microglial chemotaxis and propagation of a pro-inflammatory signa (El Khoury &

Luster, 2008). Microglia can also use CCL-4 to enhance the recruitment of additional microglia (Rock et al., 2004). Thus, it may be involved in microglial recruitment or propagation of the chemotherapy-induced pro-inflammatory signal within the CNS. Further research is needed to fully understand the role of increased CCL-4 after chemotherapy.

To evaluate the causal relationship between elevated proinflammatory cytokines and altered affective behavior following chemotherapy, we administered minocycline beginning several days prior to chemotherapy

112 treatment. Minocycline is a lipid soluble broad-spectrum tetracycline antibiotic, most commonly used to treat patients with acne and Lyme disease. It readily crosses the BBB and is known to suppress microglial activation and associated increase in pro-inflammatory cytokines within the CNS through blockade of NF- kappa B nuclear translocation (Plane, Shen, Pleasure, & Deng, 2010). Thus, it was hypothesized that minocycline would ameliorate chemotherapy-induced increase in microglial activation and proinflammatory cytokines, subsequently ameliorating chemotherapy-induced depressive-like and anxiety-like behavior.

Indeed, minocycline administration did concurrently reduce chemotherapy- induced depressive-like behavior, anxiety-like behavior, and gene expression of

IL-6. However, there was not an effect of minocycline on protein expression of pro-inflammatory cytokines or microglial activation markers in the brain (data not shown) three days after AC treatment. It may be that the primary effects of doxorubicin and cyclophosphamide on microglia and proinflammatory cytokines may subside before our analysis at 72 hours. Future research needs to assess the effects of AC on microglial activation and cytokine expression at earlier time points. Pre-clinical studies report an increase in TNFα three hours after treatment with doxorubicin (Tangpong et al., 2006). Furthermore, determining cell death within the brain at various time points, including 72 hours, will also assist in determining the effects on neurons that may have already occurred by 72 hours after treatment. It also would be beneficial to assess the effects of doxorubicin and cyclophosphamide individually to determine whether both drugs contribute

113 equally to the resulting neuroinflammation and behavioral deficits. The effect of minocycline on depressive-like behavior and IL-6 gene expression was similar for both oral and central administration of minocycline, further reinforcing the hypothesis that the effect of chemotherapy on behavior and inflammation is centrally mediated. Furthermore, oral administration of minocycline is readily available in clinical settings.

These data suggest that doxorubicin and cyclophosphamide may induce depressive-like and anxiety-like behavior. Furthermore this effect may be mediated by an increase in pro-inflammatory cytokines released from activated microglia. More importantly, minocycline ameliorates the depressive-like and anxiety-like effects of AC. This is relevant for clinical research as minocycline is an antibiotic already prescribed and used in clinical settings. Concomitant administration of minocycline and chemotherapeutic agents may ameliorate the effects on affective behavior seen in a subset of patients treated with chemotherapy.

114

1000 *

300 *

) ) )

800

mL mL

mL 250

mL mL mL

pg/ pg/ pg/ 200 600

150

protein ( ( protein protein

protein ( protein 400

α α

α 100

CCL4 protein (pg/ (pg/ (pg/ protein protein protein CCL4 CCL4 CCL4

TNFTNF TNF 200 50

0 0 Vehicle Chemo Vehicle Chemo

Figure 3.1 Chemotherapy increases protein expression of TNFα and CCL4 in the brain 72 hours after treatment. Proinflammatory proteins markers tumor necrosis factor alpha (TNFα) and chemokine (C-C motif) ligand 4 (also known as macrophage inflammatory protein 1ß) were measured in whole brain homogenates using bioplex. The data are presented as mean (+SEM); an asterisk (*) indicates a significant difference between groups at p<0.05.

115

A B * C 0.8 0.6 *

0.6 0.6 0.4

0.4 0.4

6 Rel. Gene Exp. Exp. Exp. 666Gene Gene Gene Rel. Rel. Rel.

1b Rel. Gene Exp. Exp. Exp. 1b1b1bGene Gene Gene Rel. Rel. Rel.

- -

- 0.2

- - -

IL IL

0.2 IL

IL IL IL

TNFa Rel.Rel. TNFa TNFa GeneGeneExp. Exp. TNFa Rel. TNFa GeneExp. 0.2

0 0 0 Vehicle Chemo Vehicle Chemo Vehicle Chemo

Figure 3.2 Chemotherapy increases mRNA expression of TNFα and IL-6, but not IL-1β in the prefrontal cortex of the brain 72 hours after treatment. Gene expression of proinflammatory protein markers tumor necrosis factor alpha (TNFα; A), IL-6 (B), and IL-1ß (C) were measured in the prefrontal cortex using RT-PCR. The data are presented as mean (+SEM); an asterisk (*) indicates a significant difference between groups at p<0.05.

116

A B C 50 * 12 100

40 10 80 8 30 60 6 20 40 4 10

Number of floating bouts bouts bouts floating floating floating NumberNumberNumber of of of

Time floating (seconds) (seconds) TimeTimefloating floating

Time floating (seconds) Timefloating 2

Latency to float (seconds) (seconds) (seconds) float float float to to to Latency Latency Latency

0 0 0 Vehicle Chemo Vehicle Chemo Vehicle Chemo

Figure 3.3 Chemotherapy increases depressive-like behavior in BALBc mice 72 hours after treatment. Depressive-like behavior was assessed during a three minutes test using the forced swim task. The data are presented as mean (+SEM); an asterisk (*) indicates a significant difference between groups at p<0.05.

117

A B * 18 30 14

20 10

TotalTotalarmarm entries entries

TimeTimearmsarms (seconds) (seconds) open open in in 0 0 Vehicle Chemo Vehicle Chemo

Figure 3.4 Chemotherapy increases anxiety-like behavior in BalbC mice without changing the locomotor activity 72 hours after treatment. Anxiety- like behavior was assessed in a five minute task using the elevated plus maze. The data are presented as mean (+SEM); an asterisk (*) indicates a significant difference between groups at p<0.05.

118

2500

2000

1500

1000

500

Number ofcounts Number locomotor 0 Vehicle Chemo

Figure 3.5 Chemotherapy does not affect locomotor activity in BALBc mice 72 hours after treatment. Locomotor activity was assessed in a 30 minute task using the open field test. The data are presented as mean (+SEM); an asterisk (*) indicates a significant difference between groups at p<0.05.

119

A B C 60 12 80 * 50 10 60

40 8

(seconds) (seconds) 30 6 40

20 4 20

10 2

Time floating TimeTimefloating floating

Latency to float (seconds) (seconds) float float to to Latency Latency

Number of floating bouts bouts floating floating NumberNumber of of

0 0 0 Vehicle Chemo Vehicle Chemo Vehicle Chemo

Figure 3.6 Chemotherapy increases depressive like behavior in C57bl/6 mice 72 hours after treatment. Depressive-like behavior was assessed using a three minute forced swim task. The data are presented as mean (+SEM); an asterisk (*) indicates a significant difference between groups at p<0.05.

120

A B * 18 30 14

20 10

6 10

Number of total armarm NumberNumberEntries Entries total total of of Time (seconds) in open arms open Timein (seconds) Time (seconds) in open arms open Timein (seconds) 0 0 Vehicle Chemo Vehicle Chemo

Figure 3.7 Chemotherapy increases anxiety-like behavior of C57bl/6 mice without affecting locomotor activity 72 hours after treatment. Anxiety-like behavior was assessed during a five minute test using the elevated plus maze. The data are presented as mean (+SEM); an asterisk (*) indicates a significant difference between groups at p<0.05.

121

5000

4000

3000

2000

1000

NumberNumberlocomotorlocomotor of of counts counts 0 Vehicle Chemo

Figure 3.8 Chemotherapy (72 hours) does not affect locomotor activity in C57bl/6 mice 72 hours after treatment. Locomotor activity was assessed in a 30 minute task using the open field test. The data are presented as mean (+SEM); an asterisk (*) indicates a significant difference between groups at p<0.05.

122

32 0.3 1.4 * * *

1 30 0.2

0.6 28 0.1

% cd11b+ cells % cd11b+ 0.2

%cd11b+cd80+ cells %cd11b+cd80+ %cd11b+cd86+ cells %cd11b+cd86+ 0 26 0 Vehicle Chemo Vehicle Chemo Vehicle Chemo

Figure 3.9 Chemotherapy increases microglial activation markers within the brain 72 hours after treatment. Protein expression of microglial activation markers cd11b, cd80, and cd86 were assessed in whole brain homogenates using flow cytometry. The data are presented as mean (+SEM); an asterisk (*) indicates a significant difference between groups at p<0.05.

123

Cyclophosphamide Phosphamide mustard Doxorubicin

) ) )

) 100000.00

nM nM nM nM 10000.00 1000.00 100.00

10.00

Concentrations ( ( ( Concentrations Concentrations Concentrations Concentrations ( Concentrations 1.00 0.00 4.00 8.00 12.00 16.00 20.00 24.00 Time (hr) B Cyclophosphamide Phosphamide mustard

10.0000

) ) ) )

1.0000

nmol/g nmol/g nmol/g nmol/g

0.1000

0.0100

Concentrations ( ( ( ( Concentrations Concentrations Concentrations Concentrations 0.0010 0.00 0.50 1.00 1.50 2.00 2.50 3.00 3.50 4.00 Time (hr)

Figure 3.10 Expression of chemotherapeutic agents in the blood and brain at different time points after treatment. Concentrations of cyclophosphamide, doxorubicin, and phosphamide mustard metabolite were measured in the blood (A; nmol/liter) and whole brain homogenates (B; nmol/gram) using liquid chromatography tandem mass spectrometry (LC-MS/MS). The data are presented as mean (+SEM).

124

A B 50 a 18 a a,b 40 b 14 b b,c 30 c b 10 20 6

Time floating (seconds) (seconds) (seconds) (seconds) TimeTimeTimeTimefloating floating floating floating

Number of floating bouts bouts bouts bouts floating floating floating floating NumberNumberNumberNumber of of of of Number of floating bouts floating Number of 2 0 0 Vehicle Chemo Chemo Vehicle Vehicle Chemo Chemo Vehicle Water Water Minocycline Minocycline Water Water Minocycline Minocycline

C 80 a a a 60 a

Latency to float (seconds) (seconds) (seconds) (seconds) (seconds) float float float float float to to to to to Latency Latency Latency Latency Latency 0 Vehicle Chemo Chemo Vehicle Water Water Minocycline Minocycline

Figure 3.11 Minocycline (i.c.v.) ameliorates chemotherapy-induced depressive-like behavior 72 hours treatment. Depressive-like behavior was assessed using a three minute forced swim task. The data are presented as mean (+SEM); groups containing the same letter are not significantly different from one another (p>0.05).

125

60 a a a

20 b

Time in open arms (seconds) (seconds) arms arms open open inin Time Time 0 Vehicle Vehicle Minocycline Minocycline Vehicle Chemo Chemo Vehicle

Figure 3.12 Minocycline (i.c.v.) ameliorates chemotherapy-induced anxiety- like behavior 72 hours after treatment. Anxiety-like behavior was assessed from a five minute task using the elevated plus maze. The data are presented as mean (+SEM); groups containing the same letter are not significantly different from one another (p>0.05).

126

3000 a 2500 a a a 2000

1500

1000

Total locomotor countscounts TotalTotallocomotorlocomotor Total locomotor counts Totallocomotor 500

0 Vehicle Chemo Chemo Vehicle Water Water Minocycline Minocycline

Figure 3.13 Minocycline (i.c.v.) does not affect overall locomotor activity after 72 hour chemotherapy treatment. Locomotor activity was assessed from a 30 minute task using the open field test. The data are presented as mean (+SEM); groups containing the same letter are not significantly different from one another (p>0.05).

127

a 1.2

1 b b,c 0.8 c

0.6

0.4 6 rel. gene expression gene rel. 6

6 rel. gene expression gene rel. 6 0.2

- -

IL IL 0 Vehicle Chemo Chemo Vehicle Water Water Minocycline Minocycline

Figure 3.14 Minocycline (i.c.v.) ameliorates chemotherapy-induced gene expression of pro-inflammatory cytokine IL-6 in the prefrontal cortex 72 hours after treatment. Gene expression of proinflammatory marker IL-6 was measured in the pre-frontal cortex using RT-PCR. The data are presented as mean (+SEM); groups containing the same letter are not significantly different from one another (p>0.05).

128

A B 50 a 18 a a,b 40 b 14 b b,c 30 c b 10 20 6

Time floating (seconds) (seconds) (seconds) (seconds) TimeTimeTimeTimefloating floating floating floating

C 80 a a a 60 a

Figure 3.15 Minocycline (oral) ameliorates chemotherapy-induced depressive-like behavior 72 hours treatment. Depressive-like behavior was assessed using a three minute forced swim task. The data are presented as mean (+SEM); groups containing the same letter are not significantly different from one another (p>0.05).

129

3500 a 3000 a a a 2500

2000

1500

1000

500

Total locomotor counts counts counts TotalTotalTotallocomotor locomotor locomotor 0 Vehicle Chemo Chemo Vehicle Water Water Minocycline Minocycline

Figure 3.16 Minocycline (oral) does not affect overall locomotor activity after 72 hour chemotherapy treatment. Locomotor activity was assessed from a 30 minute task using the open field test. The data are presented as mean (+SEM); groups containing the same letter are not significantly different from one another (p>0.05).

130

a 1.2

1 b b,c 0.8 c

0.6

0.4 6 rel. gene expression gene rel. 6

6 rel. gene expression gene rel. 6 0.2

- -

IL IL 0 Vehicle Chemo Chemo Vehicle Water Water Minocycline Minocycline

Figure 3.17 Minocycline (oral) ameliorates chemotherapy-induced gene expression of pro-inflammatory cytokine IL-6 in the prefrontal cortex 72 hours after treatment. Gene expression of proinflammatory marker IL-6 was measured in the pre-frontal cortex using RT-PCR. The data are presented as mean (+SEM); groups containing the same letter are not significantly different from one another (p>0.05).

131

Chapter 4

The Effects of Chemotherapy on Cognition

The effects of chemotherapy on cognition have garnered much interest and more than a thousand papers on this topic have been published, many in breast cancer patients due to the relatively high survival rate. Chemotherapy- related cognitive impairment is commonly reported by BC patients (Myers, 2012).

The majority of BC studies that incorporate cognitive self-report measures indicate that women who have been treated with chemotherapy report greater cognitive deficits than BC patients who have not been treated with chemotherapy and matched healthy controls. Longitudinal studies using tests to assess cognitive changes after chemotherapy report significant deficits following treatment. Furthermore, these deficits can be detected up to fifteen years after receiving chemotherapy (Yamada, Denburg, Beglinger, & Schultz, 2010).

Although the clinical studies in breast cancer patients provide convincing evidence that a significant proportion of women undergoing chemotherapy develop persistent cognitive and affective side effects, and despite the vast literature, the mechanism is unknown. Animal models have been beneficial in that they allow a systematic and controlled study of the physiological changes in the brain associated with treatment of individual or combined chemotherapeutic agents. As discussed previously, a majority of animal research has shown that

132 chemotherapeutic agents do indeed induce deleterious cognitive effects in non- tumor models (ElBeltagy et al., 2010; Konat, Kraszpulski, James, Zhang, &

Abraham, 2008; Y. Li, V. Vijayanathan, M. E. Gulinello, & P. D. Cole, 2010;

Mondie, Vandergrift, Wilson, Gulinello, & Weber, 2010; Reiriz et al., 2006;

Winocur et al., 2012; M. Yang et al., 2010). In rodent models, cyclophosphamide impairs passive avoidance task learning, novel object recognition, and inhibitory avoidance in mice (Reiriz et al., 2006; M. Yang et al., 2010) while doxorubicin impairs inhibitory avoidance in rats (Liedke et al., 2009). Combined, these drugs impair passive avoidance and contextual fear (Konat, Kraszpulski, James,

Zhang, & Abraham, 2008).

In the past decade, a growing number of pre-clinical studies have proposed potential mechanisms associated with chemotherapeutic neurotoxicity within the CNS, but no interventional studies have attempted to establish causation. The goals of the present study are 1) to determine the cognitive effects of AC treatment using the Barnes maze, 2) to assess the morphological effects of AC on neurons and 3) to determine whether minocycline is able to reduce the reduce the chemotherapy-induced cognitive deficits. If minocycline is able to reverse the cognitive deficits, then this research could offer the first treatment of cognitive deficits in BC patients.

133

Experimental Design

Experiment 1

Sixteen Balb/C female mice were ovariectomized two weeks before the experiment. Eight mice received a single tail vein injection of doxorubicin (13.5 mg/kg) and cyclophosphamide (135 mg/kg) chemotherapeutic agents (AC); eight received saline (vehicle). Cognitive testing using the Barnes maze began two days after treatment. After each test mice were placed back in their cage for ten minutes before beginning the next test. Following the last day of testing brains were collected and prepped for PCR analysis. Although behavior was performed for six days, statistical analysis for behavior was performed on the first five days as both groups learned the task by the 5th day.

Experiment 2

Twenty Balb/C female mice were ovariectomized two weeks before the experiment. Ten mice received a single tail vein injection of AC; ten received vehicle. Brains were collected six days after treatment and placed in Golgi Cox solution (FD Rapid Golgi Stain Kit) then sectioned (100 μm) for neuronal morphology analysis.

134

Experiment 3

Forty Balb/C female mice were ovariectomized two weeks before the experiment. Beginning two days before receiving AC treatment (tail vein) and remaining until the end of the study, mice were administered either minocycline

(90 mg/kg) in their drinking water or the typical water; the water was changed every day for each animal. Cognitive testing using the Barnes maze began two days after treatment. Following the last day of testing, brains were collected and prepped for PCR analysis.

Results

Experiment 1

Learning and memory

Chemotherapy impaired spatial learning in the Barnes maze; both chemotherapy and vehicle-treated mice learned the maze as indicated by a significant decrease in latency to reach the target box (F4,52=67.3, p< .05; Figure

4.1A) and number of errors (F4,44=3.3, p< .05; Figure 4.1B) over the course of training. However, chemotherapy treated mice exhibited significantly longer latencies to reach the escape box than vehicle mice (F4,52=2.96, p< .05).

Likewise, chemotherapy treated mice exhibited significantly longer latencies to reach the escape box than pair housed mice on days 2, 3, and 4. (t13= 2.3, 2.9, and 2.5, respectively).

135

Cytokine gene expression

Chemotherapy upregulated gene expression of pro-inflammatory cytokines in the hippocampus (Figure 4.2) and prefrontal cortex (Figure 4.3) of mice after completion of the Barnes maze. In the hippocampus, relative gene expression was greater in the chemotherapy treated mice for IL-1β (Figure 4.3A; t11=3.3, p<.05) and IL-6 (Figure 4.3B; t11=4.5, p<.01) and approached significance for TNFα (Figure 4.3C; t11=1.6, p=.07). Chemotherapy also reduced prefrontal cortex gene expression of TNFα (Figure 4.4A; t10=2.8, p<.05) and approached significance for IL-1β (Figure 4.4B; t9=1.4, p=0.1).

Experiment 2

Neuronal morphology

Golgi Cox stain was used to assess neuronal dendritic spine density in the hippocampus as learning and memory is associated with an increase in spine density. Chemotherapy treatment reduced dendritic spine density in both apical and basal dendrites of the CA1 and CA3 (Figure 4.4; apical CA1: t18 = 2.76; p<.05; basal CA1: t18 = 3.27; p<.01; apical CA3: t17 = 3.18; p<.01; basal CA3: t17

= 3.8; p<.01) as well as the basal dendrites of the dentate gyrus in the hippocampus (t18 = 2.78; p<.05). There was no effect of chemotherapy on the apical dendrites in the dentate gyrus (t18 = 2.78; p>.05).

136

Experiment 3

Learning and memory

Minocycline (90 mg/kg/day) ameliorated chemotherapy-induced cognitive impairment in the Barnes maze (Figure 4.5) Repeated measures analysis reported differences between groups (F4,112=3.6, p< .05) with post-hoc analysis revealing that chemotherapy treated mice receiving water exhibited significantly longer latencies to reach the escape box than all other groups (p < .05), including the group of chemo mice receiving minocycline, which did not differ from either of the non-chemo treated groups. Likewise, chemotherapy treated mice had significantly more errors on day 3 (F3,31=6.6, p<.05) and traveled longer distances on days 3 (F3,28=7.9, p<.01;) and 4 (F3,28=5.1, p<.01) than the other three groups although the overall repeated measures analyses were not significantly different

(errors: F4,112=0.99, p>.05; distance: F4,112=1.15, p>.05).

Gene expression of pro-inflammatory markers

Minocycline (90 mg/kg/day) ameliorated chemotherapy-induced gene expression of pro-inflammatory cytokine IL-6 in the prefrontal cortex (Figure 4.6A;

F3,26=4.6, p<.05). Post-hoc analysis revealed that mice receiving minocycline and chemotherapy had reduced gene expression of pro-inflammatory marker IL-6 relative to mice receiving normal drinking water and chemotherapy (p < .05). The data approached significance for IL-6 in the hippocampus (Figure 4.6B; F3,24=

1.76, p=.09) For IL-1β receptor in the prefrontal cortex there was an overall

137 difference (Figure 4.6C; F3,27= 3.69, p<.05) with post-hoc analysis revealing higher IL-1βr in the water+chemotherapy group than the water+vehicle and minocycline+vehicle groups and approached significance in comparison to minocycline chemo group (p=.06). There were no differences for IL-1β or TNF-α

(data not shown).

Discussion

Clinical findings suggest that breast cancer patients treated with chemotherapeutic agents are more likely to experience cognitive deficits than non-treated controls (Ahles et al., 2010; Collins, Mackenzie, Stewart, Bielajew, &

Verma, 2009). Furthermore, one study found that patients receiving AC-based treatment reported more cognitive deficits than those receiving other chemotherapeutic agents (Janelsins et al., 2012). Thus, I hypothesized that non- tumor bearing mice treated with doxorubicin and cyclophosphamide would also show an increase in cognitive deficits relative to vehicle treated mice. Indeed, relative to control mice, chemotherapy treated mice showed reduced learning and memory ability in the Barnes maze in that they spent more time, committed more errors, and traveled a greater distance before finding the escape box than the vehicle-treated mice. Similar to our previous findings, we hypothesized that treatment with concomitant doxorubicin and cyclophosphamide would increase proinflammatory markers within the brain. Indeed, gene expression of pro-

138 inflammatory cytokines IL-6 and IL-1β in the hippocampus and TNF- α in the prefrontal cortex were elevated in AC treated mice relative to control mice. As dendritic spine density on neurons within the hippocampus is frequently associated with learning and memory (Leuner, Falduto, & Shors, 2003; Moser,

Trommald, & Andersen, 1994), it was hypothesized that AC treatment would result in a decrease in dendritic spine density within the hippocampus. Indeed, mice treated with chemotherapy had fewer dendritic spines within the CA1, CA3, and dentate gyrus of the hippocampus than control mice. As previously discussed, minocycline has an anti-inflammatory effect within the CNS and was successful in ameliorating chemotherapy-induced depressive-like and anxiety- like behavior. Thus, it was hypothesized that minocycline would ameliorate chemotherapy-induced increase in proinflammatory cytokines, subsequently ameliorating the associated cognitive deficits. Indeed, minocycline administration did ameliorate chemotherapy-induced increase in IL-6 gene expression in the prefrontal cortex and approached significance in ameliorating IL-6 increase in the hippocampus and IL-1βreceptor in the prefrontal cortex. As increased IL-1β receptor expression enhances the effects of IL-1β, suppressing the expression of

IL-1β receptor reduces the downstream effects initiated by IL-1β binding to its receptor, thus dampening the overall effect of IL-1β. Minocycline also ameliorated chemotherapy-induced learning and memory deficits in the Barnes maze. There was no minocycline effect on IL-1β or TNF- α. As clinical studies have reported a reduction in gray matter after chemotherapy treatment (de Ruiter

139 et al., 2011; Inagaki et al., 2007), and pre-clinical studies have reported reduction in neurogenisis (ElBeltagy et al., 2010; Janelsins et al., 2010), future studies need to assess the effects of AC on cell death and neurogenesis, as minocycline may affect one or both of these processes.

These data provide evidence in support of the hypothesis that doxorubicin and cyclophosphamide induce deficits in learning and memory. Furthermore this effect is mediated by an increase in pro-inflammatory cytokines. Most importantly, minocycline ameliorates the chemotherapy-induced deficits in learning and memory. This is relevant for clinical research as currently there is no treatment for cognitive deficits in cancer patients and many breast cancer patients experience “chemo-fog”. Concomitant administration of minocycline and chemotherapeutic agents may ameliorate the effects on cognitive ability reported in a subset of patients treated with chemotherapeutic agents. As minocycline is an antibiotic already prescribed and used in clinical settings, a clinical trial should be relatively easy to begin and is needed to test the effect of minocycline in breast cancer patients getting ready to undergo chemotherapy treatment. If minocycline is able to prevent the onset of cognitive deficits with chemotherapy, then this research could offer the first treatment of cognitive deficits in BC patients.

140

A B * 125 * 5 100 Vehicle Vehicle 4 Chemo Chemo 75 3 50

Latency Latency (s) # of errors 25 1

0 0 T1 T2 T3 T4 T5 T6 T1 T2 T3 T4 T5 T6 Trial Day Trial day

Figure 4.1 Chemotherapy impairs learning and memory in the Barnes maze 2-7 days after treatment. Learning and memory was assessed using latency to the hole and number of errors over six days using the Barnes maze; an asterisk (*) indicates a significant difference between groups at p<0.05.

141

A B C 1.4 * 2 1.2 * 1.5 1 0.8 1

0.6 1b Rel. Gene Exp 1bGene Rel.

- 0.4

6 Rel. Gene Exp. 6Gene Rel. -

IL 0.5 TNFa Rel. TNFa GeneExp. 0.2 IL 0 0 0 Vehicle Chemo Vehicle Chemo Vehicle Chemo

Figure 4.2 Chemotherapy increases mRNA expression of IL-1b and IL-6 in the hippocampus 7 days hours after treatment. Gene expression of proinflammatory protein markers IL-1ß (A), IL-6 (B), and TNFα (C), were measured in the hippocampus using RT-PCR. The data are presented as mean (+SEM); an asterisk (*) indicates a significant difference between groups at p<0.05.

142

A B 1.8 * 1.6 1.4 1.2 1 0.8 0.6

0.4

1b Rel. Gene Exp. Gene Rel. 1b -

0.2 IL TNFa Rel. Gene Exp. Gene TNFaRel. 0 0 Vehicle Chemo Vehicle Chemo

Figure 4.3 Chemotherapy increases mRNA expression of TNFα in the prefrontal cortex 7 days after treatment. Gene expression of proinflammatory protein markers TNFα (A) and and IL-1ß (B), were measured in the prefrontal cortex using RT-PCR. The data are presented as mean (+SEM); an asterisk (*) indicates a significant difference between groups at p<0.05.

143

A B

1.4 CA3 Basal

1.2 * * * * * 1

0.8

0.6

0.4 spine density (per um) (per spinedensity spine density (per um) (per spinedensity 0.2

ChemoChemo ChemoChemo ChemoChemo

Vehicle Vehicle Vehicle Vehicle Vehicle Vehicle

Vehicle Vehicle Vehicle Vehicle

Vehicle Vehicle Vehicle Apical CA1 Basal CA1 Apical CA3 Basal CA3 Apical DG Basal DG Vehicle Chemo

Figure 4.4 Chemotherapy decreases dendritic spine density in the hippocampus 3 days after treatment. Dendritic spine density was quantified in the CA1, CA3, and dentate gyrus regions of the hippocampus following golgi-cox staining. The data are presented as mean (+SEM); an asterisk (*) indicates a significant difference between groups at p<0.05.

144

A * Vehicle Water 100 Chemo Water Chemo Mino Vehicle Mino

Latency (sec) (sec) (sec) Latency Latency Latency

0 T1 T2 T3 T4 T5 Trial Day B C * 2.5 * 10.0 2.0 7.5 * 1.5

5.0 1.0

###errors errors errors of of of

2.5

Distance (meters) (meters) Distance Distance Distance (meters) Distance 0.5

0.0 0.0 T1 T2 T3 T4 T5 T1 T2 T3 T4 T5 Trial Day Trial Day

Figure 4.5 Minocycline ameliorates chemotherapy-induced cognitive impairment in the Barnes maze 2-6 days after treatment. Learning and memory was assessed using latency to the hole, number of errors, and distance to the hole over six days using the Barnes maze; an asterisk (*) indicates a significant difference between groups at p<0.05.

145

4 A B a 2 a

3 a 1.5 a b b b a 2 1

1 0.5

6 Rel. Gene Expression Expression Expression Gene Gene Gene Rel. Rel. Rel. 6 6 6

- - -

IL IL IL

IL IL IL 0 0 Vehicle Chemo Chemo Vehicle Vehicle Chemo Chemo Vehicle Water Water Minocycline Minocycline Water Water Minocycline Minocycline C 2.5 a

2 a,b b 1.5 b

1b Rel. Gene Expression Expression Expression Gene Gene Gene Rel. Rel. Rel. 1b 1b 1b

1b Rel. Gene Expression Gene Rel. 1b 0.5

- - - -

IL IL IL IL

0 Vehicle Chemo Chemo Vehicle Water Water Minocycline Minocycline

Figure 4.6 Figure 3.17 Minocycline ameliorates chemotherapy-induced gene expression of pro-inflammatory cytokine IL-6 in the prefrontal cortex 6 days after treatment. Gene expression of proinflammatory marker IL-6 and IL- 1β were measured in the pre-frontal cortex (A and C) and hippocampus (B) using RT-PCR. The data are presented as mean (+SEM); groups containing the same letter are not significantly different from one another (p>0.05).

146

Chapter 5

Conclusion

Breast cancer is the most frequently diagnosed cancer in women; one in eight women in the U.S. will be diagnosed with this disease during their lifetime

(Howlader N, 2011). Due to improved detection and treatment advances, the overall 5-year and 10-year survival rates for breast cancer are now 90% and

84.5% respectively (Howlader N, 2011). Thus, there is a growing population of breast cancer survivors for whom persistent side effects of treatment could impinge upon quality of life. Three of the most common side effects reported in breast cancer patients receiving chemotherapy are cognitive deficits, depression, and anxiety. These symptoms affect quality of life and are associated with decreased survival, increased recurrence rates, and increased tumor growth in breast cancer patients, although it is not clear if a causal relationship exist between depression and these outcome measures (Falagas et al., 2007;

Goodwin, Zhang, & Ostir, 2004; Khan, Amatya, Pallant, & Rajapaksa, 2012;

Meyer, Sinnott, & Seed, 2003; Redeker, Lev, & Ruggiero, 2000).

Despite the high incidence of cognitive and affective symptoms, the mechanisms through which chemotherapy may be affecting brain function remains a matter of speculation; prior to the research described in this thesis, no causal mechanism had been established in either humans or other animals.

Furthermore, there are currently no standard pharmacological treatments for

147 chemotherapy-induced cognitive dysfunction and treatment of depression using conventional pharmaceutical methods is limited by potential interference with the effectiveness of endocrine therapy (Kelly et al., 2010). My studies not only demonstrate that both the cognitive and affective side effects of chemotherapy are related to neuroinflammation, but provide strong preclinical evidence that minocycline, an approved tetracycline analogue with good tolerability, can alleviate both of these side effects.

The goals of this dissertation were to assess the effects of doxorubicin and cyclophosphamide on cognitive and affective behavior, investigate a possible mechanism, and test the effects of minocycline on suppressing chemo- induced changes in behavior using a mouse model.

Based on both clinical and pre-clinical findings, it was hypothesized, in

Chapter 3, that a single dose of AC would increase depressive-like and anxiety- like behavior. Indeed, a single treatment [75% human equivalent dose (HED)] of doxorubicin and cyclophosphamide increased depressive-like and anxiety-like behavior. Given the scarcity of the literature, both clinical and pre-clinical, associating depression and anxiety with chemotherapy, these findings are beneficial, specifically in a non-tumor model, suggesting that the chemotherapeutic agents may increase depressive and anxiogenic symptoms.

In Chapter 4, I tested the hypothesis that AC treatment would induce learning and memory deficits. Indeed, mice receiving a single dose of concomitant doxorubicin and cyclophosphamide (75% HED) showed impaired

148 cognitive ability in that they spent more time, had more errors, and traveled a greater distance before reaching the target hide box. Using a novel cognitive test for this field, the Barnes maze, these findings support findings from previous rodent models suggesting that chemotherapy may induce cognitive deficits in non-tumor bearing rodents (ElBeltagy et al., 2010; Reiriz et al., 2006; M. Yang et al., 2010).

Both clinical and pre-clinical studies have found doxorubicin to be associated with an increase in pro-inflammatory markers in both the blood and the brain (Aluise et al., 2011; Janelsins et al., 2012; Kang, Chess-Williams,

Anoopkumar-Dukie, & McDermott, 2013; Tangpong et al., 2011). Thus, it was hypothesized that treatment with AC would increase pro-inflammatory markers within the brain. Indeed, gene expression of IL-6, TNF-α, and IL-1b were higher in chemotherapy-treated mice relative to controls in the hippocampus and prefrontal cortex. Furthermore, levels of pro-inflammatory cytokines TNF- α and

CCL4 were higher in chemotherapy-treated mice than controls.

Numerous pre-clinical studies, using a wide range of models, report that upregulation of microglia within the brain increases pro-inflammatory cytokines which can induce sickness response and depressive-like behavior (Dantzer,

O'Connor, Freund, Johnson, & Kelley, 2008; Neigh et al., 2009; Norman et al.,

149 report an association between depression and increased serum concentrations of IL-6 and inflammatory mediator NF-ĸB (Jehn et al.; Musselman et al., 2001;

Torres et al., 2013). Thus, it was hypothesized that microglia mediate the effects of chemotherapy on depressive-like and anxiety-like behavior. Although we did detect an increase in microglial activation markers CD80 and CD86 three days after treatment, more extensive studies on enriched microglia, with and without

LPS stimulation, showed no chemotherapy effect on microglial activation. To better understand the potential role that microglia may have in mediating chemotherapy and behavior, future studies will need to assess microglial activation at earlier time points, as it is possible that microglia do become activated after chemotherapy treatment, but then return to a quiescent state within 72 hours after treatment. It is also possible that astrocytes may play a role in mediating these effects, which needs to be addressed in future studies.

Minocycline is a lipid soluble broad-spectrum tetracycline antibiotic that readily crosses the BBB and is known to suppress microglial activation and associated increase in pro-inflammatory cytokines within the CNS through blockade of NF-kappa B nuclear translocation (Plane, Shen, Pleasure, & Deng,

2010). Thus, it was hypothesized that minocycline would ameliorate chemotherapy-induced increase in microglial activation and proinflammatory cytokines, subsequently ameliorating chemotherapy-induced depressive-like and anxiety-like behavior. Indeed, minocycline administration did ameliorate chemotherapy-induced depressive-like behavior, anxiety-like behavior, and gene

150 expression of IL-6. However, there was not an effect of minocycline on protein expression of pro-inflammatory cytokines or microglial activation markers in the brain. Further research is needed to understand the role of microglia and inflammation in mediating the effects of chemotherapy on behavior. Specifically, kinetic studies would be beneficial in assessing the time course of inflammation from the periphery to the brain, followed by the cell-specific effects within the brain.

Although the mechanism is not fully understood, the main purpose of this proposal was to assess the possibility of minocycline being able to suppress or ameliorate the effects of AC on behavior. Administration of minocycline directly into the brain via intracerebral cannulation was effective in ameliorating the chemo-induced increases in depressive-like and anxiety-like behavior. Oral administration of minocycline was also effective at reversing the effects of chemotherapy on depressive-like behavior. There was not an effect of minocycline on anxiety-like behavior, however, when administered orally. It may be that a higher dose of minocycline is needed to have an effect on anxiety-like behavior; further research is needed to test this hypothesis. These findings are clinically significant as oral route is the easiest and most convenient mode of treatment administration and i.c.v. administration would not be very plausible in treating humans.

Minocycline was also successful in ameliorating the effects of chemotherapy on cognition. Mice receiving a single AC injection receiving normal

151 drinking water spent more time searching for the hide box and had more errors on day 3 and traveled a greater distance on days three and four than the other groups. Chemotherapy treated mice receiving minocycline in their drinking water performed significantly better on the Barnes maze than the non-minocycline chemotherapy group. Furthermore, the time spent searching for the hide box was almost identical on each test day for the minocycline treated chemo group as the two control groups, suggesting that minocycline may be able to completely abolish the cognitive deficits induced by chemotherapy. In is also important to note the minocycline itself did not have an effect on behavior, as noted by the similarity in behavior between the water vehicle group and the minocycline vehicle group.

As clinical studies have reported a reduction in gray matter after chemotherapy treatment (de Ruiter et al., 2011; Inagaki et al., 2007), and pre- clinical studies have reported reduction in neurogenisis (ElBeltagy et al., 2010;

Janelsins et al., 2010), future studies need to assess the effects of AC on cell death and neurogenesis, as minocycline may affect one or both of these. As breast cancer patients receive multiple doses of chemotherapeutic drugs, further pre-clinical research is also needed to determine if multiple treatments results in more severe cognitive and affective side effects.

My data suggest that a single administration of doxorubicin and cyclophosphamide is enough to alter cognitive and affective behavior.

Furthermore, this effect is likely to be mediated by an increase in pro-

152 inflammatory cytokines released from activated microglia. Minocycline is a lipid soluble broad-spectrum tetracycline antibiotic, most commonly used to treat patients with acne and Lyme disease. It readily crosses the BBB and is known to suppress microglial activation and associated increase in pro-inflammatory cytokines within the CNS through blockade of NF-kappa B nuclear translocation

(Plane, Shen, Pleasure, & Deng, 2010). My studies indicate that minocycline ameliorates the cognitive, depressive-like and anxiety-like effects of AC in rodents. This is relevant for clinical research as many breast cancer patients experience chemotherapy-related neuropsychological deficits and there are currently no standard pharmacological treatments for chemotherapy-induced cognitive dysfunction. Furthermore, treatment of depression using conventional pharmaceutical methods is limited by potential interference with the effectiveness of endocrine therapy (ET; Kelly et al., 2010). These side effects can compromise quality of life by amplifying the physical symptoms and functional impairments experienced by the women (Fann et al., 2008). In addition, untreated depression and concern about side effects can contribute to poor treatment adherence

(Colleoni et al., 2000).

Concomitant administration of minocycline and chemotherapeutic agents may ameliorate the effects on cognitive ability seen in a subset of patients treated with chemotherapeutic agents. As minocycline is an antibiotic already prescribed and used in clinical settings, a clinical trial should be relatively easy to begin and is needed to test the effect of minocycline in breast cancer patients

153 getting ready to undergo chemotherapy treatment. If minocycline is able to reverse the neuropsychological deficits in clinical trials, then this research could offer the first treatment of cognitive deficits in BC patients and be a safe option for treating chemo-induced depression and anxiety.

154

References

Abraham, J., Haut, M. W., Moran, M. T., Filburn, S., Lemiuex, S., & Kuwabara, H. (2008). Adjuvant chemotherapy for breast cancer: effects on cerebral white matter seen in diffusion tensor imaging. Clin Breast Cancer, 8(1), 88-91.

Ahles, T. A., & Saykin, A. J. (2007). Candidate mechanisms for chemotherapy- induced cognitive changes. Nat Rev Cancer, 7(3), 192-201.

Ahles, T. A., Saykin, A. J., Furstenberg, C. T., Cole, B., Mott, L. A., Skalla, K., et al. (2002). Neuropsychologic impact of standard-dose systemic chemotherapy in long-term survivors of breast cancer and lymphoma. J Clin Oncol, 20(2), 485-493.

Ahles, T. A., Saykin, A. J., McDonald, B. C., Li, Y., Furstenberg, C. T., Hanscom, B. S., et al. (2010). Longitudinal assessment of cognitive changes associated with adjuvant treatment for breast cancer: impact of age and cognitive reserve. J Clin Oncol, 28(29), 4434-4440.

Aletti, G. D., Nordquist, D., Hartmann, L., Gallenberg, M., Long, H. J., & Cliby, W. A. (2010). From randomized trial to practice: single institution experience using the GOG 172 i.p. chemotherapy regimen for ovarian cancer. Ann Oncol, 21(9), 1772-1778.

Aluise, C. D., Miriyala, S., Noel, T., Sultana, R., Jungsuwadee, P., Taylor, T. J., et al. (2011). 2-Mercaptoethane sulfonate prevents doxorubicin-induced plasma protein oxidation and TNF-alpha release: implications for the reactive oxygen species-mediated mechanisms of chemobrain. Free Radic Biol Med, 50(11), 1630-1638.

Aluise, C. D., Sultana, R., Tangpong, J., Vore, M., St Clair, D., Moscow, J. A., et al. (2010). Chemo brain (chemo fog) as a potential side effect of doxorubicin administration: role of cytokine-induced, oxidative/nitrosative stress in cognitive dysfunction. Adv Exp Med Biol, 678, 147-156.

155

Alvarez, A., Cacabelos, R., Sanpedro, C., Garcia-Fantini, M., & Aleixandre, M. (2007). Serum TNF-alpha levels are increased and correlate negatively with free IGF-I in Alzheimer disease. Neurobiol Aging, 28(4), 533-536.

American Cancer Society. Cancer Facts & Figures 2013. Atlanta: American Cancer Society; 2013.

American Cancer Society. Breast Cancer Facts & Figures 2011-2012. Atlanta: American Cancer Society, Inc.

Anwar, M. J., Pillai, K. K., Khanam, R., Akhtar, M., & Vohora, D. (2011). Effect of alprazolam on anxiety and cardiomyopathy induced by doxorubicin in mice. Fundam Clin Pharmacol, 26(3), 356-362.

Armstrong, D. K., Bundy, B., Wenzel, L., Huang, H. Q., Baergen, R., Lele, S., et al. (2006). Intraperitoneal cisplatin and paclitaxel in ovarian cancer. N Engl J Med, 354(1), 34-43.

Avis, N. E., Levine, B., Naughton, M. J., Case, D. L., Naftalis, E., & Van Zee, K. J. (2012).Explaining age-related differences in depression following breast cancer diagnosis and treatment. Breast Cancer Res Treat, 136(2), 581- 591.

Bell, I. R., Edman, J. S., Selhub, J., Morrow, F. D., Marby, D. W., Kayne, H. L., et al. (1992). Plasma homocysteine in vascular disease and in nonvascular dementia of depressed elderly people. Acta Psychiatr Scand, 86(5), 386- 390.

Bender, C. M., Sereika, S. M., Berga, S. L., Vogel, V. G., Brufsky, A. M., Paraska, K. K., et al. (2006). Cognitive impairment associated with adjuvant therapy in breast cancer. Psychooncology, 15(5), 422-430.

Berent-Spillson, A., Persad, C. C., Love, T., Sowers, M., Randolph, J. F., Zubieta, J. K., et al. Hormonal environment affects cognition independent of age during the menopause transition. J Clin Endocrinol Metab, 97(9), E1686-1694.

156

Bhatia, A. L., Manda, K., Patni, S., & Sharma, A. L. (2006). Prophylactic action of linseed (Linum usitatissimum) oil against cyclophosphamide-induced oxidative stress in mouse brain. J Med Food, 9(2), 261-264.

Bines, J., Oleske, D. M., & Cobleigh, M. A. (1996). Ovarian function in premenopausal women treated with adjuvant chemotherapy for breast cancer. J Clin Oncol, 14(5), 1718-1729.

Binkhorst, L., Mathijssen, R. H., van Herk-Sukel, M. P., Bannink, M., Jager, A., Wiemer, E. A., et al. (2013). Unjustified prescribing of CYP2D6 inhibiting SSRIs in women treated with tamoxifen. Breast Cancer Res Treat, 139(3), 923-929.

Bjorneklett, H. G., Lindemalm, C., Rosenblad, A., Ojutkangas, M. L., Letocha, H., Strang, P., et al. (2012). A randomised controlled trial of support group intervention after breast cancer treatment: results on anxiety and depression. Acta Oncol, 51(2), 198-207.

Bower, J. E., Ganz, P. A., Desmond, K. A., Rowland, J. H., Meyerowitz, B. E., & Belin, T. R. (2000). Fatigue in breast cancer survivors: occurrence, correlates, and impact on quality of life. J Clin Oncol, 18(4), 743-753.

Bower, J. E., Ganz, P. A., Irwin, M. R., Castellon, S., Arevalo, J., & Cole, S. W. (2013). Cytokine genetic variations and fatigue among patients with breast cancer. J Clin Oncol, 31(13), 1656-1661.

Bower, J. E., Ganz, P. A., Irwin, M. R., Kwan, L., Breen, E. C., & Cole, S. W. (2011). Inflammation and behavioral symptoms after breast cancer treatment: do fatigue, depression, and sleep disturbance share a common underlying mechanism? J Clin Oncol, 29(26), 3517-3522.

Boyes, A. W., Girgis, A., D'Este, C. A., Zucca, A. C., Lecathelinais, C., & Carey, M. L. (2013). Prevalence and predictors of the short-term trajectory of anxiety and depression in the first year after a cancer diagnosis: a population-based longitudinal study. J Clin Oncol, 31(21), 2724-2729.

157

Boyette-Davis, J. A., & Fuchs, P. N. (2009). Differential effects of paclitaxel treatment on cognitive functioning and mechanical sensitivity. Neurosci Lett, 453(3), 170-174.

Briones, T. L., & Woods, J. (2011). Chemotherapy-induced cognitive impairment is associated with decreases in cell proliferation and histone modifications. BMC Neurosci, 12, 124.

Brown, M. S., Simon, J. H., Stemmer, S. M., Stears, J. C., Scherzinger, A., Cagnoni, P. J., et al. (1995). MR and proton spectroscopy of white matter disease induced by high-dose chemotherapy with bone marrow transplant in advanced breast carcinoma. AJNR Am J Neuroradiol, 16(10), 2013- 2020.

Brown, M. S., Stemmer, S. M., Simon, J. H., Stears, J. C., Jones, R. B., Cagnoni, P. J., et al. (1998). White matter disease induced by high-dose chemotherapy: longitudinal study with MR imaging and proton spectroscopy. AJNR Am J Neuroradiol, 19(2), 217-221.

Brufsky, A. M., Mayer, M., Rugo, H. S., Kaufman, P. A., Tan-Chiu, E., Tripathy, D., et al. (2011). Central nervous system metastases in patients with HER2-positive metastatic breast cancer: incidence, treatment, and survival in patients from registHER. Clin Cancer Res, 17(14), 4834-4843.

Burgess, C., Cornelius, V., Love, S., Graham, J., Richards, M., & Ramirez, A. (2005). Depression and anxiety in women with early breast cancer: five year observational cohort study. Bmj, 330(7493), 702.

Caplette-Gingras, A., & Savard, J. (2008). Depression in women with metastatic breast cancer: a review of the literature. Palliat Support Care, 6(4), 377- 387.

Christie, L. A., Acharya, M. M., Parihar, V. K., Nguyen, A., Martirosian, V., & Limoli, C. L. (2012). Impaired cognitive function and hippocampal neurogenesis following cancer chemotherapy. Clin Cancer Res, 18(7), 1954-1965.

158

Cimprich, B., Reuter-Lorenz, P., Nelson, J., Clark, P. M., Therrien, B., Normolle, D., et al. (2010). Prechemotherapy alterations in brain function in women with breast cancer. J Clin Exp Neuropsychol, 32(3), 324-331.

Cimprich, B., So, H., Ronis, D. L., & Trask, C. (2005). Pre-treatment factors related to cognitive functioning in women newly diagnosed with breast cancer. Psychooncology, 14(1), 70-78.

Cleeland, C. S., Bennett, G. J., Dantzer, R., Dougherty, P. M., Dunn, A. J., Meyers, C. A., et al. (2003). Are the symptoms of cancer and cancer treatment due to a shared biologic mechanism? A cytokine-immunologic model of cancer symptoms. Cancer, 97(11), 2919-2925.

Colleoni, M., Mandala, M., Peruzzotti, G., Robertson, C., Bredart, A., & Goldhirsch, A. (2000). Depression and degree of acceptance of adjuvant cytotoxic drugs. Lancet, 356(9238), 1326-1327.

Collins, B., Mackenzie, J., Stewart, A., Bielajew, C., & Verma, S. (2009). Cognitive effects of chemotherapy in post-menopausal breast cancer patients 1 year after treatment. Psychooncology, 18(2), 134-143.

Collins, B., MacKenzie, J., Tasca, G. A., Scherling, C., & Smith, A. (2012). Cognitive effects of chemotherapy in breast cancer patients: a dose- response study. Psychooncology, 22(7), 1517-1527.

Corona, A. W., Fenn, A. M., & Godbout, J. P. Cognitive and behavioral consequences of impaired immunoregulation in aging. J Neuroimmune Pharmacol, 7(1), 7-23.

Dantzer, R., O'Connor, J. C., Freund, G. G., Johnson, R. W., & Kelley, K. W. (2008). From inflammation to sickness and depression: when the immune system subjugates the brain. Nat Rev Neurosci, 9(1), 46-56. de Jong, N., Courtens, A. M., Abu-Saad, H. H., & Schouten, H. C. (2002). Fatigue in patients with breast cancer receiving adjuvant chemotherapy: a review of the literature. Cancer Nurs, 25(4), 283-297; quiz 298-289.

159 de Ruiter, M. B., Reneman, L., Boogerd, W., Veltman, D. J., Caan, M., Douaud, G., et al. (2011). Late effects of high-dose adjuvant chemotherapy on white and gray matter in breast cancer survivors: converging results from multimodal magnetic resonance imaging. Hum Brain Mapp, 33(12), 2971- 2983.

Del Mastro, L., Catzeddu, T., Boni, L., Bell, C., Sertoli, M. R., Bighin, C., et al. (2006). Prevention of chemotherapy-induced menopause by temporary ovarian suppression with goserelin in young, early breast cancer patients. Ann Oncol, 17(1), 74-78.

Deprez, S., Amant, F., Smeets, A., Peeters, R., Leemans, A., Van Hecke, W., et al. (2012). Longitudinal assessment of chemotherapy-induced structural changes in cerebral white matter and its correlation with impaired cognitive functioning. J Clin Oncol, 30(3), 274-281.

Deprez, S., Amant, F., Yigit, R., Porke, K., Verhoeven, J., Van den Stock, J., et al. (2011). Chemotherapy-induced structural changes in cerebral white matter and its correlation with impaired cognitive functioning in breast cancer patients. Hum Brain Mapp, 32(3), 480-493.

Dietrich, J., Han, R., Yang, Y., Mayer-Proschel, M., & Noble, M. (2006). CNS progenitor cells and oligodendrocytes are targets of chemotherapeutic agents in vitro and in vivo. J Biol, 5(7), 22.

Donovan, K. A., Small, B. J., Andrykowski, M. A., Schmitt, F. A., Munster, P., & Jacobsen, P. B. (2005). Cognitive functioning after adjuvant chemotherapy and/or radiotherapy for early-stage breast carcinoma. Cancer, 104(11), 2499-2507.

Dunlop, B. W., & Nemeroff, C. B. (2007). The role of dopamine in the pathophysiology of depression. Arch Gen Psychiatry, 64(3), 327-337.

Eijkenboom, M., & Van Der Staay, F. J. (1999). Spatial learning deficits in rats after injection of vincristine into the dorsal hippocampus. Neuroscience, 91(4), 1299-1313.

160

El Khoury, J., & Luster, A. D. (2008). Mechanisms of microglia accumulation in Alzheimer's disease: therapeutic implications. Trends Pharmacol Sci, 29(12), 626-632.

ElBeltagy, M., Mustafa, S., Umka, J., Lyons, L., Salman, A., Chur-yoe, G. T., et al. (2010). Fluoxetine improves the memory deficits caused by the chemotherapy agent 5-fluorouracil. Behav Brain Res, 208(1), 112-117.

Falagas, M. E., Zarkadoulia, E. A., Ioannidou, E. N., Peppas, G., Christodoulou, C., & Rafailidis, P. I. (2007). The effect of psychosocial factors on breast cancer outcome: a systematic review. Breast Cancer Res, 9(4), R44.

Falleti, M. G., Sanfilippo, A., Maruff, P., Weih, L., & Phillips, K. A. (2005). The nature and severity of cognitive impairment associated with adjuvant chemotherapy in women with breast cancer: a meta-analysis of the current literature. Brain Cogn, 59(1), 60-70.

Fann, J. R., Thomas-Rich, A. M., Katon, W. J., Cowley, D., Pepping, M., McGregor, B. A., et al. (2008). Major depression after breast cancer: a review of epidemiology and treatment. Gen Hosp Psychiatry, 30(2), 112- 126.

Fardell, J. E., Vardy, J., Logge, W., & Johnston, I. (2010). Single high dose treatment with methotrexate causes long-lasting cognitive dysfunction in laboratory rodents. Pharmacol Biochem Behav, 97(2), 333-339.

Fardell, J. E., Vardy, J., Shah, J. D., & Johnston, I. N. (2012). Cognitive impairments caused by oxaliplatin and 5-fluorouracil chemotherapy are ameliorated by physical activity. Psychopharmacology (Berl), 220(1), 183- 193.

Fiorentino, L., & Ancoli-Israel, S. (2006). Insomnia and its treatment in women with breast cancer. Sleep Med Rev, 10(6), 419-429.

Fiorentino, L., Rissling, M., Liu, L., & Ancoli-Israel, S. (2012). The Symptom Cluster of Sleep, Fatigue and Depressive Symptoms in Breast Cancer Patients: Severity of the Problem and Treatment Options. Drug Discov Today Dis Models, 8(4), 167-173. 161

Fishel, M. L., Vasko, M. R., & Kelley, M. R. (2007). DNA repair in neurons: so if they don't divide what's to repair? Mutat Res, 614(1-2), 24-36.

Fremouw, T., Fessler, C. L., Ferguson, R. J., & Burguete, Y. (2012). Preserved learning and memory in mice following chemotherapy: 5-Fluorouracil and doxorubicin single agent treatment, doxorubicin-cyclophosphamide combination treatment. Behav Brain Res, 226(1), 154-162.

Gandal, M. J., Ehrlichman, R. S., Rudnick, N. D., & Siegel, S. J. (2008). A novel electrophysiological model of chemotherapy-induced cognitive impairments in mice. Neuroscience, 157(1), 95-104.

Ganz, P. A., Kwan, L., Castellon, S. A., Oppenheim, A., Bower, J. E., Silverman, D. H., et al. (2013). Cognitive complaints after breast cancer treatments: examining the relationship with neuropsychological test performance. J Natl Cancer Inst, 105(11), 791-801.

Gerstner, E. R., & Fine, R. L. (2007). Increased permeability of the blood-brain barrier to chemotherapy in metastatic brain tumors: establishing a treatment paradigm. J Clin Oncol, 25(16), 2306-2312.

Goodwin, J. S., Zhang, D. D., & Ostir, G. V. (2004). Effect of depression on diagnosis, treatment, and survival of older women with breast cancer. J Am Geriatr Soc, 52(1), 106-111.

Greenberg, D. B. (2004). Barriers to the treatment of depression in cancer patients. J Natl Cancer Inst Monogr(32), 127-135.

Griffin, A. C., & Whitacre, C. C. (1991). Sex and strain differences in the circadian rhythm fluctuation of endocrine and immune function in the rat: implications for rodent models of autoimmune disease. J Neuroimmunol, 35(1-3), 53-64.

Grimm, S. A. (2011). Treatment of brain metastases: chemotherapy. Curr Oncol Rep, 14(1), 85-90.

162

Hack, T. F., Pickles, T., Ruether, J. D., Weir, L., Bultz, B. D., Mackey, J., et al. Predictors of distress and quality of life in patients undergoing cancer therapy: impact of treatment type and decisional role. Psychooncology, 19(6), 606-616.

Han, R., Yang, Y. M., Dietrich, J., Luebke, A., Mayer-Proschel, M., & Noble, M. (2008). Systemic 5-fluorouracil treatment causes a syndrome of delayed myelin destruction in the central nervous system. J Biol, 7(4), 12.

Hankey, B. F., Miller, B., Curtis, R., & Kosary, C. (1994). Trends in breast cancer in younger women in contrast to older women. J Natl Cancer Inst Monogr(16), 7-14.

Hascoet, M., & Bourin, M. (2009). The Mouse Light-Dark Box Test. In Mood and Anxiety Related Phenotypes in Mice Neuromethods (Vol. 42, pp. 197- 223): Humana Press.

Helal, G. K., Aleisa, A. M., Helal, O. K., Al-Rejaie, S. S., Al-Yahya, A. A., Al- Majed, A. A., et al. (2009). Metallothionein induction reduces caspase-3 activity and TNFalpha levels with preservation of cognitive function and intact hippocampal neurons in carmustine-treated rats. Oxid Med Cell Longev, 2(1), 26-35.

Henry, C. J., Huang, Y., Wynne, A. M., & Godbout, J. P. (2009). Peripheral lipopolysaccharide (LPS) challenge promotes microglial hyperactivity in aged mice that is associated with exaggerated induction of both pro- inflammatory IL-1beta and anti-inflammatory IL-10 cytokines. Brain Behav Immun, 23(3), 309-317.

Hermelink, K., Untch, M., Lux, M. P., Kreienberg, R., Beck, T., Bauerfeind, I., et al. (2007). Cognitive function during neoadjuvant chemotherapy for breast cancer: results of a prospective, multicenter, longitudinal study. Cancer, 109(9), 1905-1913.

Hirschfeld, R. M. (2001). The Comorbidity of Major Depression and Anxiety Disorders: Recognition and Management in Primary Care. Prim Care Companion J Clin Psychiatry, 3(6), 244-254.

163

Hodgson, K. D., Hutchinson, A. D., Wilson, C. J., & Nettelbeck, T. (2013). A meta-analysis of the effects of chemotherapy on cognition in patients with cancer. Cancer Treat Rev, 39(3), 297-304.

Hopwood, P., Howell, A., & Maguire, P. (1991). Psychiatric morbidity in patients with advanced cancer of the breast: prevalence measured by two self- rating questionnaires. Br J Cancer, 64(2), 349-352.

Hotopf, M., Chidgey, J., Addington-Hall, J., & Ly, K. L. (2002). Depression in advanced disease: a systematic review Part 1. Prevalence and case finding. Palliat Med, 16(2), 81-97.

Howlader N, N. A., Krapcho M, et al. . (2011). SEER Cancer Statistics Review, 1975-2008. Bethesda, MD: National Cancer Institute.

Hurria, A., Rosen, C., Hudis, C., Zuckerman, E., Panageas, K. S., Lachs, M. S., et al. (2006). Cognitive function of older patients receiving adjuvant chemotherapy for breast cancer: a pilot prospective longitudinal study. J Am Geriatr Soc, 54(6), 925-931.

Hwang, S. Y., Chang, S. J., & Park, B. W. (2013). Does chemotherapy really affect the quality of life of women with breast cancer? J Breast Cancer, 16(2), 229-235.

Inagaki, M., Yoshikawa, E., Matsuoka, Y., Sugawara, Y., Nakano, T., Akechi, T., et al. (2007). Smaller regional volumes of brain gray and white matter demonstrated in breast cancer survivors exposed to adjuvant chemotherapy. Cancer, 109(1), 146-156.

Janelsins, M. C., Mustian, K. M., Palesh, O. G., Mohile, S. G., Peppone, L. J., Sprod, L. K., et al. (2012). Differential expression of cytokines in breast cancer patients receiving different chemotherapies: implications for cognitive impairment research. Support Care Cancer, 20(4), 831-839.

Janelsins, M. C., Roscoe, J. A., Berg, M. J., Thompson, B. D., Gallagher, M. J., Morrow, G. R., et al. (2010). IGF-1 partially restores chemotherapy- induced reductions in neural cell proliferation in adult C57BL/6 mice. Cancer Invest, 28(5), 544-553. 164

Jansen, C. E., Dodd, M. J., Miaskowski, C. A., Dowling, G. A., & Kramer, J. (2008). Preliminary results of a longitudinal study of changes in cognitive function in breast cancer patients undergoing chemotherapy with doxorubicin and cyclophosphamide. Psychooncology, 17(12), 1189-1195.

Jehn, C. F., Flath, B., Strux, A., Krebs, M., Possinger, K., Pezzutto, A., et al. Influence of age, performance status, cancer activity, and IL-6 on anxiety and depression in patients with metastatic breast cancer. Breast Cancer Res Treat, 136(3), 789-794.

Jenkins, V., Shilling, V., Deutsch, G., Bloomfield, D., Morris, R., Allan, S., et al. (2006). A 3-year prospective study of the effects of adjuvant treatments on cognition in women with early stage breast cancer. Br J Cancer, 94(6), 828-834.

Jenkins, V., Shilling, V., Fallowfield, L., Howell, A., & Hutton, S. (2004). Does hormone therapy for the treatment of breast cancer have a detrimental effect on memory and cognition? A pilot study. Psychooncology, 13(1), 61- 66.

Jim, H. S., Phillips, K. M., Chait, S., Faul, L. A., Popa, M. A., Lee, Y. H., et al. (2012). Meta-analysis of cognitive functioning in breast cancer survivors previously treated with standard-dose chemotherapy. J Clin Oncol, 30(29), 3578-3587.

Johnston, A., Gudjonsson, J. E., Sigmundsdottir, H., Ludviksson, B. R., & Valdimarsson, H. (2005). The anti-inflammatory action of methotrexate is not mediated by lymphocyte apoptosis, but by the suppression of activation and adhesion molecules. Clin Immunol, 114(2), 154-163.

Jonasson, Z. (2005). Meta-analysis of sex differences in rodent models of learning and memory: a review of behavioral and biological data. Neurosci Biobehav Rev, 28(8), 811-825.

Joshi, G., Hardas, S., Sultana, R., St Clair, D. K., Vore, M., & Butterfield, D. A. (2007). Glutathione elevation by gamma-glutamyl cysteine ethyl ester as a potential therapeutic strategy for preventing oxidative stress in brain mediated by in vivo administration of adriamycin: Implication for chemobrain. J Neurosci Res, 85(3), 497-503. 165

Kang, S. H., Chess-Williams, R., Anoopkumar-Dukie, S., & McDermott, C. (2013). Induction of inflammatory cytokines and alteration of urothelial ATP, acetylcholine and prostaglandin E2 release by doxorubicin. Eur J Pharmacol, 700(1-3), 102-109.

Keller, J. N., Schmitt, F. A., Scheff, S. W., Ding, Q., Chen, Q., Butterfield, D. A., et al. (2005). Evidence of increased oxidative damage in subjects with mild cognitive impairment. Neurology, 64(7), 1152-1156.

Kelly, C. M., Juurlink, D. N., Gomes, T., Duong-Hua, M., Pritchard, K. I., Austin, P. C., et al. (2010). Selective serotonin reuptake inhibitors and breast cancer mortality in women receiving tamoxifen: a population based cohort study. Bmj, 340, c693.

Kesler, S., Janelsins, M., Koovakkattu, D., Palesh, O., Mustian, K., Morrow, G., et al. (2013). Reduced hippocampal volume and verbal memory performance associated with interleukin-6 and tumor necrosis factor-alpha levels in chemotherapy-treated breast cancer survivors. Brain Behav Immun, 30 Suppl, S109-116.

Kesler, S. R., Bennett, F. C., Mahaffey, M. L., & Spiegel, D. (2009). Regional brain activation during verbal declarative memory in metastatic breast cancer. Clin Cancer Res, 15(21), 6665-6673.

Kesler, S. R., Kent, J. S., & O'Hara, R. (2011). Prefrontal cortex and executive function impairments in primary breast cancer. Arch Neurol, 68(11), 1447- 1453.

Kesler, S. R., Wefel, J. S., Hosseini, S. M., Cheung, M., Watson, C. L., & Hoeft, F. (2013). Default mode network connectivity distinguishes chemotherapy- treated breast cancer survivors from controls. Proc Natl Acad Sci U S A, 110(28), 11600-11605.

Khan, F., Amatya, B., Pallant, J. F., & Rajapaksa, I. (2012). Factors associated with long-term functional outcomes and psychological sequelae in women after breast cancer. Breast, 21(3), 314-320.

166

Kissane, D. W., Bloch, S., Smith, G. C., Miach, P., Clarke, D. M., Ikin, J., et al. (2003). Cognitive-existential group psychotherapy for women with primary breast cancer: a randomised controlled trial. Psychooncology, 12(6), 532- 546.

Kissane, D. W., Grabsch, B., Love, A., Clarke, D. M., Bloch, S., & Smith, G. C. (2004). Psychiatric disorder in women with early stage and advanced breast cancer: a comparative analysis. Aust N Z J Psychiatry, 38(5), 320- 326.

Kittas, C., & Henry, L. (1980). Effect of sex hormones on the response of mice to infection with Toxoplasma gondii. Br J Exp Pathol, 61(6), 590-600.

Klein, S. L., & Nelson, R. J. (1998). Adaptive immune responses are linked to the mating system of arvicoline rodents. Am Nat, 151(1), 59-67.

Knobf, M. T. (2007). Psychosocial responses in breast cancer survivors. Semin Oncol Nurs, 23(1), 71-83.

Komatsu, H., Hayashi, N., Suzuki, K., Yagasaki, K., Iioka, Y., Neumann, J., et al. (2012). Guided self-help for prevention of depression and anxiety in women with breast cancer. ISRN Nurs, 2012, 716367.

Konat, G. W., Kraszpulski, M., James, I., Zhang, H. T., & Abraham, J. (2008). Cognitive dysfunction induced by chronic administration of common cancer chemotherapeutics in rats. Metab Brain Dis, 23(3), 325-333.

Koopman, C., Nouriani, B., Erickson, V., Anupindi, R., Butler, L. D., Bachmann, M. H., et al. (2002). Sleep disturbances in women with metastatic breast cancer. Breast J, 8(6), 362-370.

Koppelmans, V., de Ruiter, M. B., van der Lijn, F., Boogerd, W., Seynaeve, C., van der Lugt, A., et al. (2012). Global and focal brain volume in long-term breast cancer survivors exposed to adjuvant chemotherapy. Breast Cancer Res Treat, 132(3), 1099-1106.

167

Koros, C., & Kitraki, E. (2009). Neurofilament isoform alterations in the rat cerebellum following cytosine arabinoside administration. Toxicol Lett, 189(3), 215-218.

Kujjo, L. L., Chang, E. A., Pereira, R. J., Dhar, S., Marrero-Rosado, B., Sengupta, S., et al. (2011). Chemotherapy-induced late transgenerational effects in mice. PLoS One, 6(3), e17877.

Lamers, F., van Oppen, P., Comijs, H. C., Smit, J. H., Spinhoven, P., van Balkom, A. J., et al. (2011). Comorbidity patterns of anxiety and depressive disorders in a large cohort study: the Netherlands Study of Depression and Anxiety (NESDA). J Clin Psychiatry, 72(3), 341-348.

Lau, B. W., Yau, S. Y., Lee, T. M., Ching, Y. P., Tang, S. W., & So, K. F. (2009). Intracerebroventricular infusion of cytosine-arabinoside causes prepulse inhibition disruption. Neuroreport, 20(4), 371-377.

Lee, G. D., Longo, D. L., Wang, Y., Rifkind, J. M., Abdul-Raman, L., Mamczarz, J. A., et al. (2006). Transient improvement in cognitive function and synaptic plasticity in rats following cancer chemotherapy. Clin Cancer Res, 12(1), 198-205.

Lee, V. W., & Harris, D. C. (2011). Adriamycin nephropathy: a model of focal segmental glomerulosclerosis. Nephrology (Carlton), 16(1), 30-38.

Leuner, B., Falduto, J., & Shors, T. J. (2003). Associative memory formation increases the observation of dendritic spines in the hippocampus. J Neurosci, 23(2), 659-665.

Li, C. Q., Liu, D., Huang, L., Wang, H., Zhang, J. Y., & Luo, X. G. (2008). Cytosine arabinoside treatment impairs the remote spatial memory function and induces dendritic retraction in the anterior cingulate cortex of rats. Brain Res Bull, 77(5), 237-240.

Li, Y., Vijayanathan, V., Gulinello, M., & Cole, P. D. (2010). Intrathecal methotrexate induces focal cognitive deficits and increases cerebrospinal fluid homocysteine. Pharmacol Biochem Behav, 95(4), 428-433.

168

Li, Y., Vijayanathan, V., Gulinello, M. E., & Cole, P. D. (2010). Systemic methotrexate induces spatial memory deficits and depletes cerebrospinal fluid folate in rats. Pharmacol Biochem Behav, 94(3), 454-463.

Liedke, P. E., Reolon, G. K., Kilpp, B., Brunetto, A. L., Roesler, R., & Schwartsmann, G. (2009). Systemic administration of doxorubicin impairs aversively motivated memory in rats. Pharmacol Biochem Behav, 94(2), 239-243.

Lim, C. C., Devi, M. K., & Ang, E. (2011). Anxiety in women with breast cancer undergoing treatment: a systematic review. Int J Evid Based Healthc, 9(3), 215-235.

Liu, L., Fiorentino, L., Natarajan, L., Parker, B. A., Mills, P. J., Sadler, G. R., et al. (2009). Pre-treatment symptom cluster in breast cancer patients is associated with worse sleep, fatigue and depression during chemotherapy. Psychooncology, 18(2), 187-194.

Long, J. M., Lee, G. D., Kelley-Bell, B., Spangler, E. L., Perez, E. J., Longo, D. L., et al. (2011). Preserved learning and memory following 5-fluorouracil and cyclophosphamide treatment in rats. Pharmacol Biochem Behav, 100(1), 205-211.

Lyons, L., Elbeltagy, M., Bennett, G., & Wigmore, P. (2011). The effects of cyclophosphamide on hippocampal cell proliferation and spatial working memory in rat. PLoS One, 6(6), e21445.

Lyons, L., ElBeltagy, M., Bennett, G., & Wigmore, P. (2012). Fluoxetine counteracts the cognitive and cellular effects of 5-fluorouracil in the rat hippocampus by a mechanism of prevention rather than recovery. PLoS One, 7(1), e30010.

Lyons, L., ElBeltagy, M., Umka, J., Markwick, R., Startin, C., Bennett, G., et al. (2011). Fluoxetine reverses the memory impairment and reduction in proliferation and survival of hippocampal cells caused by methotrexate chemotherapy. Psychopharmacology (Berl), 215(1), 105-115.

169

Macleod, J. E., DeLeo, J. A., Hickey, W. F., Ahles, T. A., Saykin, A. J., & Bucci, D. J. (2007). Cancer chemotherapy impairs contextual but not cue-specific fear memory. Behav Brain Res, 181(1), 168-172.

Madhyastha, S., Somayaji, S. N., Rao, M. S., Nalini, K., & Bairy, K. L. (2002). Hippocampal brain amines in methotrexate-induced learning and memory deficit. Can J Physiol Pharmacol, 80(11), 1076-1084.

Maes, M., Bosmans, E., De Jongh, R., Kenis, G., Vandoolaeghe, E., & Neels, H. (1997). Increased serum IL-6 and IL-1 receptor antagonist concentrations in major depression and treatment resistant depression. Cytokine, 9(11), 853-858.

Magaki, S., Mueller, C., Dickson, C., & Kirsch, W. (2007). Increased production of inflammatory cytokines in mild cognitive impairment. Exp Gerontol, 42(3), 233-240.

Manda, K., & Bhatia, A. L. (2003). Prophylactic action of melatonin against cyclophosphamide-induced oxidative stress in mice. Cell Biol Toxicol, 19(6), 367-372.

Massie, M. J. (2004). Prevalence of depression in patients with cancer. J Natl Cancer Inst Monogr(32), 57-71.

McDonald, B. C., Conroy, S. K., Ahles, T. A., West, J. D., & Saykin, A. J. (2010). Gray matter reduction associated with systemic chemotherapy for breast cancer: a prospective MRI study. Breast Cancer Res Treat, 123(3), 819- 828.

Mehlsen, M., Pedersen, A. D., Jensen, A. B., & Zachariae, R. (2009). No indications of cognitive side-effects in a prospective study of breast cancer patients receiving adjuvant chemotherapy. Psychooncology, 18(3), 248- 257.

Meyer, H. A., Sinnott, C., & Seed, P. T. (2003). Depressive symptoms in advanced cancer. Part 2. Depression over time; the role of the palliative care professional. Palliat Med, 17(7), 604-607.

170

Mikova, O., Yakimova, R., Bosmans, E., Kenis, G., & Maes, M. (2001). Increased serum tumor necrosis factor alpha concentrations in major depression and multiple sclerosis. Eur Neuropsychopharmacol, 11(3), 203-208.

Miller, A. H., Maletic, V., & Raison, C. L. (2009). Inflammation and its discontents: the role of cytokines in the pathophysiology of major depression. Biol Psychiatry, 65(9), 732-741.

Mondie, C. M., Vandergrift, K. A., Wilson, C. L., Gulinello, M. E., & Weber, E. T. The chemotherapy agent, thioTEPA, yields long-term impairment of hippocampal cell proliferation and memory deficits but not depression- related behaviors in mice. Behav Brain Res, 209(1), 66-72.

Mondie, C. M., Vandergrift, K. A., Wilson, C. L., Gulinello, M. E., & Weber, E. T. (2010). The chemotherapy agent, thioTEPA, yields long-term impairment of hippocampal cell proliferation and memory deficits but not depression- related behaviors in mice. Behav Brain Res, 209(1), 66-72.

Moser, M. B., Trommald, M., & Andersen, P. (1994). An increase in dendritic spine density on hippocampal CA1 pyramidal cells following spatial learning in adult rats suggests the formation of new synapses. Proc Natl Acad Sci U S A, 91(26), 12673-12675.

Musselman, D. L., Miller, A. H., Porter, M. R., Manatunga, A., Gao, F., Penna, S., et al. (2001). Higher than normal plasma interleukin-6 concentrations in cancer patients with depression: preliminary findings. Am J Psychiatry, 158(8), 1252-1257.

Mustafa, S., Walker, A., Bennett, G., & Wigmore, P. M. (2008). 5-Fluorouracil chemotherapy affects spatial working memory and newborn neurons in the adult rat hippocampus. Eur J Neurosci, 28(2), 323-330.

Myers, J. S. (2012). Chemotherapy-related cognitive impairment: the breast cancer experience. Oncol Nurs Forum, 39(1), E31-40.

Nava-Castro, K., Hernandez-Bello, R., Muniz-Hernandez, S., Camacho-Arroyo, I., & Morales-Montor, J. (2012). Sex steroids, immune system, and parasitic infections: facts and hypotheses. Ann N Y Acad Sci, 1262, 16-26. 171

Neigh, G. N., Karelina, K., Glasper, E. R., Bowers, S. L., Zhang, N., Popovich, P. G., et al. (2009). Anxiety after cardiac arrest/cardiopulmonary resuscitation: exacerbated by stress and prevented by minocycline. Stroke, 40(11), 3601-3607.

Nelson, W. L., & Suls, J. (2013). New Approaches to Understand Cognitive Changes Associated With Chemotherapy for Non-Central Nervous System Tumors. J Pain Symptom Manage.

Niiya, M., Niiya, K., Kiguchi, T., Shibakura, M., Asaumi, N., Shinagawa, K., et al. (2003). Induction of TNF-alpha, uPA, IL-8 and MCP-1 by doxorubicin in human lung carcinoma cells. Cancer Chemother Pharmacol, 52(5), 391- 398.

Norman, G. J., Karelina, K., Zhang, N., Walton, J. C., Morris, J. S., & Devries, A. C. (2010). Stress and IL-1beta contribute to the development of depressive-like behavior following peripheral nerve injury. Mol Psychiatry, 15(4), 404-414.

O'Connor, J. C., Andre, C., Wang, Y., Lawson, M. A., Szegedi, S. S., Lestage, J., et al. (2009). Interferon-gamma and tumor necrosis factor-alpha mediate the upregulation of indoleamine 2,3-dioxygenase and the induction of depressive-like behavior in mice in response to bacillus Calmette-Guerin. J Neurosci, 29(13), 4200-4209.

Oboh, G., Akomolafe, T. L., Adefegha, S. A., & Adetuyi, A. O. (2012). Attenuation of cyclophosphamide-induced neurotoxicity in rat by yellow dye extract from root of Brimstone tree (Morinda lucida). Exp Toxicol Pathol, 64(6), 591-596.

Oboh, G., & Ogunruku, O. O. (2010). Cyclophosphamide-induced oxidative stress in brain: protective effect of hot short pepper (Capsicum frutescens L. var. abbreviatum). Exp Toxicol Pathol, 62(3), 227-233.

Olsen, N. J., & Kovacs, W. J. (1996). Gonadal steroids and immunity. Endocr Rev, 17(4), 369-384.

172

Oz, E., & Ilhan, M. N. (2006). Effects of melatonin in reducing the toxic effects of doxorubicin. Mol Cell Biochem, 286(1-2), 11-15.

Palumbo, M. L., Canzobre, M. C., Pascuan, C. G., Rios, H., Wald, M., & Genaro, A. M. (2010). Stress induced cognitive deficit is differentially modulated in BALB/c and C57Bl/6 mice: correlation with Th1/Th2 balance after stress exposure. J Neuroimmunol, 218(1-2), 12-20.

Penson, R. T., Kronish, K., Duan, Z., Feller, A. J., Stark, P., Cook, S. E., et al. (2000). Cytokines IL-1beta, IL-2, IL-6, IL-8, MCP-1, GM-CSF and TNFalpha in patients with epithelial ovarian cancer and their relationship to treatment with paclitaxel. Int J Gynecol Cancer, 10(1), 33-41.

Plane, J. M., Shen, Y., Pleasure, D. E., & Deng, W. (2010). Prospects for minocycline neuroprotection. Arch Neurol, 67(12), 1442-1448.

Pomykala, K. L., Ganz, P. A., Bower, J. E., Kwan, L., Castellon, S. A., Mallam, S., et al. (2013). The association between pro-inflammatory cytokines, regional cerebral metabolism, and cognitive complaints following adjuvant chemotherapy for breast cancer. Brain Imaging Behav.

Porsolt, R. D., Le Pichon, M., & Jalfre, M. (1977). Depression: a new animal model sensitive to antidepressant treatments. Nature, 266(5604), 730- 732.

Pusztai, L., Mendoza, T. R., Reuben, J. M., Martinez, M. M., Willey, J. S., Lara, J., et al. (2004). Changes in plasma levels of inflammatory cytokines in response to paclitaxel chemotherapy. Cytokine, 25(3), 94-102.

Quesnel, C., Savard, J., & Ivers, H. (2009). Cognitive impairments associated with breast cancer treatments: results from a longitudinal study. Breast Cancer Res Treat, 116(1), 113-123.

Rajamani, R., Muthuvel, A., Senthilvelan, M., & Sheeladevi, R. (2006). Oxidative stress induced by methotrexate alone and in the presence of methanol in discrete regions of the rodent brain, retina and optic nerve. Toxicol Lett, 165(3), 265-273.

173

Reagan-Shaw, S., Nihal, M., & Ahmad, N. (2008). Dose translation from animal to human studies revisited. Faseb J, 22(3), 659-661.

Redeker, N. S., Lev, E. L., & Ruggiero, J. (2000). Insomnia, fatigue, anxiety, depression, and quality of life of cancer patients undergoing chemotherapy. Sch Inq Nurs Pract, 14(4), 275-290; discussion 291-278.

Reece, J. C., Chan, Y. F., Herbert, J., Gralow, J., & Fann, J. R. (2013). Course of depression, mental health service utilization and treatment preferences in women receiving chemotherapy for breast cancer. Gen Hosp Psychiatry, 35(4), 376-381.

Reiriz, A. B., Reolon, G. K., Preissler, T., Rosado, J. O., Henriques, J. A., Roesler, R., et al. (2006). Cancer chemotherapy and cognitive function in rodent models: memory impairment induced by cyclophosphamide in mice. Clin Cancer Res, 12(16), 5000; author reply 5000-5001.

Riggs, K. M., Spiro, A., 3rd, Tucker, K., & Rush, D. (1996). Relations of vitamin B-12, vitamin B-6, folate, and homocysteine to cognitive performance in the Normative Aging Study. Am J Clin Nutr, 63(3), 306-314.

Rock, R. B., Gekker, G., Hu, S., Sheng, W. S., Cheeran, M., Lokensgard, J. R., et al. (2004). Role of microglia in central nervous system infections. Clin Microbiol Rev, 17(4), 942-964, table of contents.

Roscoe, J. A., Morrow, G. R., Hickok, J. T., Mustian, K. M., Griggs, J. J., Matteson, S. E., et al. (2005). Effect of paroxetine hydrochloride (Paxil) on fatigue and depression in breast cancer patients receiving chemotherapy. Breast Cancer Res Treat, 89(3), 243-249.

Rutledge, D. M. M., C. (2003). Evidence-based symptom management. In Cancer lymptom management. Sudbury, MA: Jones and Bartlett.

Rzeski, W., Pruskil, S., Macke, A., Felderhoff-Mueser, U., Reiher, A. K., Hoerster, F., et al. (2004). Anticancer agents are potent neurotoxins in vitro and in vivo. Ann Neurol, 56(3), 351-360.

174

Sanford, S. D., Beaumont, J. L., Butt, Z., Sweet, J. J., Cella, D., & Wagner, L. I. (2013). Prospective Longitudinal Evaluation of a Symptom Cluster in Breast Cancer. J Pain Symptom Manage.

Sawyer, A. J., Piepmeier, J. M., & Saltzman, W. M. (2006). New methods for direct delivery of chemotherapy for treating brain tumors. Yale J Biol Med, 79(3-4), 141-152.

Schagen, S. B., Hamburger, H. L., Muller, M. J., Boogerd, W., & van Dam, F. S. (2001). Neurophysiological evaluation of late effects of adjuvant high-dose chemotherapy on cognitive function. J Neurooncol, 51(2), 159-165.

Schagen, S. B., Muller, M. J., Boogerd, W., Mellenbergh, G. J., & van Dam, F. S. (2006). Change in cognitive function after chemotherapy: a prospective longitudinal study in breast cancer patients. J Natl Cancer Inst, 98(23), 1742-1745.

Schagen, S. B., van Dam, F. S., Muller, M. J., Boogerd, W., Lindeboom, J., & Bruning, P. F. (1999). Cognitive deficits after postoperative adjuvant chemotherapy for breast carcinoma. Cancer, 85(3), 640-650.

Scherling, C., Collins, B., Mackenzie, J., Bielajew, C., & Smith, A. (2011). Pre- chemotherapy differences in visuospatial working memory in breast cancer patients compared to controls: an FMRI study. Front Hum Neurosci, 5, 122.

Scherwath, A., Mehnert, A., Schleimer, B., Schirmer, L., Fehlauer, F., Kreienberg, R., et al. (2006). Neuropsychological function in high-risk breast cancer survivors after stem-cell supported high-dose therapy versus standard-dose chemotherapy: evaluation of long-term treatment effects. Ann Oncol, 17(3), 415-423.

Seigers, R., Pourtau, L., Schagen, S. B., van Dam, F. S., Koolhaas, J. M., Konsman, J. P., et al. (2010). Inhibition of hippocampal cell proliferation by methotrexate in rats is not potentiated by the presence of a tumor. Brain Res Bull, 81(4-5), 472-476.

Seigers, R., Schagen, S. B., Beerling, W., Boogerd, W., van Tellingen, O., van Dam, F. S., et al. (2008). Long-lasting suppression of hippocampal cell 175

proliferation and impaired cognitive performance by methotrexate in the rat. Behav Brain Res, 186(2), 168-175.

Seigers, R., Schagen, S. B., Coppens, C. M., van der Most, P. J., van Dam, F. S., Koolhaas, J. M., et al. (2009). Methotrexate decreases hippocampal cell proliferation and induces memory deficits in rats. Behav Brain Res, 201(2), 279-284.

Seigers, R., Schagen, S. B., Van Tellingen, O., & Dietrich, J. (2013). Chemotherapy-related cognitive dysfunction: current animal studies and future directions. Brain Imaging Behav.

Shilling, V., Jenkins, V., Morris, R., Deutsch, G., & Bloomfield, D. (2005). The effects of adjuvant chemotherapy on cognition in women with breast cancer--preliminary results of an observational longitudinal study. Breast, 14(2), 142-150.

Shilling, V., Jenkins, V., & Trapala, I. S. (2006). The (mis)classification of chemofog--methodological inconsistencies in the investigation of cognitive impairment after chemotherapy. Breast Cancer Res Treat, 95(2), 125-129.

Silberfarb, P. M., Philibert, D., & Levine, P. M. (1980). Psychosocial aspects of neoplastic disease: II. Affective and cognitive effects of chemotherapy in cancer patients. Am J Psychiatry, 137(5), 597-601.

Silverman, D. H., Dy, C. J., Castellon, S. A., Lai, J., Pio, B. S., Abraham, L., et al. (2007). Altered frontocortical, cerebellar, and basal ganglia activity in adjuvant-treated breast cancer survivors 5-10 years after chemotherapy. Breast Cancer Res Treat, 103(3), 303-311.

Silverstein, F. S., & Johnston, M. V. (1986). A model of methotrexate encephalopathy: neurotransmitter and pathologic abnormalities. J Child Neurol, 1(4), 351-357.

Snyder, J. S., Choe, J. S., Clifford, M. A., Jeurling, S. I., Hurley, P., Brown, A., et al. (2009). Adult-born hippocampal neurons are more numerous, faster maturing, and more involved in behavior in rats than in mice. J Neurosci, 29(46), 14484-14495. 176

So, W. K., Marsh, G., Ling, W. M., Leung, F. Y., Lo, J. C., Yeung, M., et al. (2009). The symptom cluster of fatigue, pain, anxiety, and depression and the effect on the quality of life of women receiving treatment for breast cancer: a multicenter study. Oncol Nurs Forum, 36(4), E205-214.

So, W. K., Marsh, G., Ling, W. M., Leung, F. Y., Lo, J. C., Yeung, M., et al. (2010). Anxiety, depression and quality of life among Chinese breast cancer patients during adjuvant therapy. Eur J Oncol Nurs, 14(1), 17-22.

Stewart, A., Bielajew, C., Collins, B., Parkinson, M., & Tomiak, E. (2006). A meta- analysis of the neuropsychological effects of adjuvant chemotherapy treatment in women treated for breast cancer. Clin Neuropsychol, 20(1), 76-89.

Stock, H. S., Rosellini, R. A., Abrahamsen, G. C., McCaffrey, R. J., & Ruckdeschel, J. C. (1995). Methotrexate does not interfere with an appetitive Pavlovian conditioning task in Sprague-Dawley rats. Physiol Behav, 58(5), 969-973.

Tangpong, J., Cole, M. P., Sultana, R., Estus, S., Vore, M., St Clair, W., et al. (2007). Adriamycin-mediated nitration of manganese superoxide dismutase in the central nervous system: insight into the mechanism of chemobrain. J Neurochem, 100(1), 191-201.

Tangpong, J., Cole, M. P., Sultana, R., Joshi, G., Estus, S., Vore, M., et al. (2006). Adriamycin-induced, TNF-alpha-mediated central nervous system toxicity. Neurobiol Dis, 23(1), 127-139.

Tangpong, J., Miriyala, S., Noel, T., Sinthupibulyakit, C., Jungsuwadee, P., & St Clair, D. K. (2011). Doxorubicin-induced central nervous system toxicity and protection by xanthone derivative of Garcinia mangostana. Neuroscience, 175, 292-299.

Torres, M. A., Pace, T. W., Liu, T., Felger, J. C., Mister, D., Doho, G. H., et al. (2013). Predictors of depression in breast cancer patients treated with radiation: role of prior chemotherapy and nuclear factor kappa B. Cancer, 119(11), 1951-1959.

177

Trask, P. C., Esper, P., Riba, M., & Redman, B. (2000). Psychiatric side effects of interferon therapy: prevalence, proposed mechanisms, and future directions. J Clin Oncol, 18(11), 2316-2326.

Tsavaris, N. C. M. P. D., Kosmas, C., Vadiaka, M., Kanelopoulos, P., & Boulamatsis, D. (2002). Immune changes in patients with advanced breast cancer undergoing chemotherapy with taxanes. British journal of cancer, 87(1), 21-27.

Vahdaninia, M., Omidvari, S., & Montazeri, A. (2010). What do predict anxiety and depression in breast cancer patients? A follow-up study. Soc Psychiatry Psychiatr Epidemiol, 45(3), 355-361. van den Beuken-van Everdingen, M. H., de Rijke, J. M., Kessels, A. G., Schouten, H. C., van Kleef, M., & Patijn, J. (2009). Quality of life and non- pain symptoms in patients with cancer. J Pain Symptom Manage, 38(2), 216-233.

Vardy, J., Rourke, S., & Tannock, I. F. (2007). Evaluation of cognitive function associated with chemotherapy: a review of published studies and recommendations for future research. J Clin Oncol, 25(17), 2455-2463.

Vearncombe, K. J., Rolfe, M., Wright, M., Pachana, N. A., Andrew, B., & Beadle, G. (2009). Predictors of cognitive decline after chemotherapy in breast cancer patients. J Int Neuropsychol Soc, 15(6), 951-962.

Verhaeghen, P., & Salthouse, T. A. (1997). Meta-analyses of age-cognition relations in adulthood: estimates of linear and nonlinear age effects and structural models. Psychol Bull, 122(3), 231-249.

Vijayanathan, V., Gulinello, M., Ali, N., & Cole, P. D. (2011). Persistent cognitive deficits, induced by intrathecal methotrexate, are associated with elevated CSF concentrations of excitotoxic glutamate analogs and can be reversed by an NMDA antagonist. Behav Brain Res, 225(2), 491-497.

Wall, P. M., & Messier, C. (2001). Methodological and conceptual issues in the use of the elevated plus-maze as a psychological measurement

178

instrument of animal anxiety-like behavior. Neurosci Biobehav Rev, 25(3), 275-286.

Wang, X. S., Shi, Q., Williams, L. A., Mao, L., Cleeland, C. S., Komaki, R. R., et al. (2010). Inflammatory cytokines are associated with the development of symptom burden in patients with NSCLC undergoing concurrent chemoradiation therapy. Brain Behav Immun, 24(6), 968-974.

Wefel, J. S., Lenzi, R., Theriault, R., Buzdar, A. U., Cruickshank, S., & Meyers, C. A. (2004). 'Chemobrain' in breast carcinoma?: a prologue. Cancer, 101(3), 466-475.

Wefel, J. S., Lenzi, R., Theriault, R. L., Davis, R. N., & Meyers, C. A. (2004). The cognitive sequelae of standard-dose adjuvant chemotherapy in women with breast carcinoma: results of a prospective, randomized, longitudinal trial. Cancer, 100(11), 2292-2299.

Wefel, J. S., Saleeba, A. K., Buzdar, A. U., & Meyers, C. A. (2010). Acute and late onset cognitive dysfunction associated with chemotherapy in women with breast cancer. Cancer, 116(14), 3348-3356.

Wefel, J. S., Vardy, J., Ahles, T., & Schagen, S. B. (2011). International Cognition and Cancer Task Force recommendations to harmonise studies of cognitive function in patients with cancer. Lancet Oncol, 12(7), 703-708.

Weiss, H. D., Walker, M. D., & Wiernik, P. H. (1974). Neurotoxicity of commonly used antineoplastic agents (first of two parts). N Engl J Med, 291(2), 75- 81.

Wenzel, L. B., Huang, H. Q., Armstrong, D. K., Walker, J. L., & Cella, D. (2007). Health-related quality of life during and after intraperitoneal versus intravenous chemotherapy for optimally debulked ovarian cancer: a Gynecologic Oncology Group Study. J Clin Oncol, 25(4), 437-443.

Winocur, G., Binns, M. A., & Tannock, I. (2011). Donepezil reduces cognitive impairment associated with anti-cancer drugs in a mouse model. Neuropharmacology, 61(8), 1222-1228.

179

Winocur, G., Henkelman, M., Wojtowicz, J. M., Zhang, H., Binns, M. A., & Tannock, I. F. (2012). The effects of chemotherapy on cognitive function in a mouse model: a prospective study. Clin Cancer Res, 18(11), 3112-3121.

Winocur, G., Vardy, J., Binns, M. A., Kerr, L., & Tannock, I. (2006). The effects of the anti-cancer drugs, methotrexate and 5-fluorouracil, on cognitive function in mice. Pharmacol Biochem Behav, 85(1), 66-75.

Wohleb, E. S., Hanke, M. L., Corona, A. W., Powell, N. D., Stiner, L. M., Bailey, M. T., et al. beta-Adrenergic receptor antagonism prevents anxiety-like behavior and microglial reactivity induced by repeated social defeat. J Neurosci, 31(17), 6277-6288.

Wynne, A. M., Henry, C. J., & Godbout, J. P. (2009). Immune and behavioral consequences of microglial reactivity in the aged brain. Integr Comp Biol, 49(3), 254-266.

Yamada, T. H., Denburg, N. L., Beglinger, L. J., & Schultz, S. K. (2010). Neuropsychological outcomes of older breast cancer survivors: cognitive features ten or more years after chemotherapy. J Neuropsychiatry Clin Neurosci, 22(1), 48-54.

Yang, I. V., Wade, C. M., Kang, H. M., Alper, S., Rutledge, H., Lackford, B., et al. (2009). Identification of novel genes that mediate innate immunity using inbred mice. Genetics, 183(4), 1535-1544.

Yang, M., Kim, J. S., Kim, J., Jang, S., Kim, S. H., Kim, J. C., et al. (2012). Acute treatment with methotrexate induces hippocampal dysfunction in a mouse model of breast cancer. Brain Res Bull, 89(1-2), 50-56.

Yang, M., Kim, J. S., Kim, J., Kim, S. H., Kim, J. C., Kim, J., et al. (2011). Neurotoxicity of methotrexate to hippocampal cells in vivo and in vitro. Biochem Pharmacol, 82(1), 72-80.

Yang, M., Kim, J. S., Song, M. S., Kim, S. H., Kang, S. S., Bae, C. S., et al. (2010). Cyclophosphamide impairs hippocampus-dependent learning and memory in adult mice: Possible involvement of hippocampal neurogenesis

180 in chemotherapy-induced memory deficits. Neurobiol Learn Mem, 93(4), 487-494.

181