Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Neuroimmunol. Author manuscript; available in PMC 2017 December 15.
Published in final edited form as:
PMCID: PMC5181109

Relationship of Systemic Cytokine Concentrations to Cognitive Function over Two Years in Women with Early Stage Breast Cancer

Debra E. Lyon, RN, PhD, Kirbo Endowed Chair, Ronald Cohen, PhD, Professor and Director, Huaihou Chen, PhD, Assistant Professor, Debra L. Kelly, PhD, RN, OCN, Assistant Professor, Nancy L. McCain, PhD, RN, FAAN, Professor Emerita, Angela Starkweather, PhD, RN, ACNP-BC, CNRN, FAAN, Professor and Director, Jamie Sturgill, PhD, Assistant Professor and Director, and Colleen K. Jackson-Cook, PhD, Professor and Director


Cancer and its treatment are frequently associated with cancer-related cognitive impairment (CRCI). While CRCI has been linked to chemotherapy, there is increasing evidence that the condition may start prior to treatment and for some, remain unresolved after active treatment and into survivorship. Although the pathophysiology of the condition is complex, alterations in systemic cytokines, signaling molecules activated in response to infection or injury that trigger inflammation, are a possible mechanism linked to cognitive dysfunction in breast cancer and other conditions. Given the conflicting results in the literature, the lack of focus on domain-specific cognitive testing, and the need for a longer time period given the multiple modalities of standard treatments for early-stage breast cancer, this longitudinal study was conducted to address these gaps.


We assessed 75 women with early-stage breast cancer at five points over two years, starting prior to the initial chemotherapy through 24 months after chemotherapy initiation. Measures included a validated computerized evaluation of domain-specific cognitive functioning and a 17-plex panel of plasma cytokines. Linear mixed-effects models were applied to test the relationships of clinical variables and cytokine concentrations to each cognitive domain. Results: Levels and patterns of cytokine concentrations varied over time: six of the 17 cytokines (IL-6, IL-12, IL-17, G-CSF, MIPS-1β, and MCP-1) had the most variability. Some cytokine levels (e.g., IL-6) increased during chemotherapy but then decreased subsequently, while others (e.g., IL-17) consistently declined from baseline over time. There were multiple relationships among cytokines and cognition, which varied over time. At baseline, elevated concentrations of G-CSF and reduced concentrations of IL-17 were associated with faster psychomotor speed. At the second time-point (prior to the mid-chemotherapy), multiple cytokines had significant associations with psychomotor speed, complex attention, executive function, verbal memory, cognitive flexibility, composite memory and visual memory. Six months after chemotherapy initiation and at the one-year point, there were multiple, significant relationships among cytokines and multiple cognitive. At two years, fewer significant relationships were noted; however, lower concentrations of IL-7, a hematopoietic cytokine, were associated with better psychomotor speed, complex attention, and memory (composite, verbal and visual). MCP-1 was inversely associated with psychomotor speed and complex attention and higher levels of MIP-1β were related to better complex attention.


Levels and patterns of cytokines changed over time and demonstrated associations with domain-specific cognitive functioning that varied over time. The observed associations between cytokines and cognitive performance provides evidence that not only prototypical cytokines (i.e. IL-6, TNF-α, and IL1-β) but also cytokines from multiple classes may contribute to the inflammatory environment that is associated with cognitive dysfunction. Future studies to better delineate the cytokine changes, both individually and in networks, are needed to precisely assess a mechanistic link between cytokines and cognitive function in women receiving treatments for breast cancer.


Cancer-related cognitive impairment (CRCI) has been the subject of considerable research inquiry, with most research focused on post-chemotherapy CRCI in women with breast cancer. The pathophysiology of CRC, although still unclear, is thought to involve the activation of cytokines, signaling molecules that mediate and regulate immunity, inflammation, and hematopoiesis. The link between inflammation and the development and progression of cancer is well-documented (Balkwill et al., 2005). Higher levels of circulating proinflammatory cytokines and their receptors have been found in individuals with cancer prior to treatment, during chemotherapy, and in survivorship. In addition to chemotherapy, cytokine interactions may be affected by other cancer treatments, including radiation therapy (Bentzen, 2006) and hormonal therapies, (Collins et al., 2009, Lee et al., 2016) as well as multiple host factors including older age, (Hurria et al., 2006, Mandelblatt et al., 2016) menopausal status and symptoms, (Miura et al., 2016) and adiposity (Hartman et al., 2015). Although cytokines have been implicated in breast cancer development and progression for many years (Dethlefsen et al., 2013), the understanding of the multiple roles of cytokines in the central nervous system has been appreciated only in the more recent past (Maier, Goehler et al. 1998; Wilson, 2002). It is now well-accepted that cytokines can cross the blood brain barrier by active transport mechanisms in the choroid plexus and circumventricular organs, and have effects on neural processing (Dantzer et al., 2008), dopamine and serotonin metabolism, neural repair and neuronal/glial cell modulation (Aluise et al., 2010). Given these mechanistic linkages, there is mounting evidence that cytokines may influence neuroinflammation and contribute to cognitive and brain dysfunction in the context of breast cancer and its treatments (Janelsins et al., 2014).

The relationships among cytokine perturbations and cognitive outcomes is emerging across many fields (Wilson et al., 2002). Distinctive cytokine signatures have been identified in multiple immune-mediated medical conditions (e.g., heart failure (Pasic et al., 2003), Alzheimer’s disease, (Swardfager et al., 2010), multiple sclerosis (Heesen et al., 2010), and Parkinson’s (Dursun et al., 2015)). In our studies, we have shown that (Cohen et al., 2011) serum cytokine concentrations are associated with attention, executive function, and learning-memory performance among adults with HIV infection. Although a relationship between systemic cytokine concentrations and CRCI has been confirmed in animal models of cancer (Janelsins et al., 2011) and in some studies of women with breast cancer undergoing chemotherapy (Briones and Woods, 2014, Galimberti et al., 2006), past studies have tended to focus on “prototypical” cytokines or have been short-term studies that have not represented the potential cytokine changes in the context of multiple cancer treatments over time. A recent study using a multiplex panel of cytokines demonstrated multiple relationships with subjective cognitive impairment and cytokine networks, but the study was truncated to the first twelve weeks of the chemotherapy phase of treatment (Cheung et al., 2015a). In addition, the investigation of cytokines as a possible biological mechanism of cognitive dysfunction has been limited by the use of multiple measures of cognition (including subjective memory complaints) and cross-sectional or short-term longitudinal measures that focus on a single aspect of treatment, such as chemotherapy, radiation, hormonal therapy or survivorship. Further, most studies have measured prototypical cytokines that limits our understanding of pro- and anti-inflammatory interactive networks. To date, there have been very few studies that have investigated cognition over time, using objective neuropsychological batteries that have started prior to the initial chemotherapy and cytokine panels that permit the measure of multiple cytokines. Given the gaps in the literature, the lack of focus on domain-specific cognitive testing, and the need for a longer time period given the multi-modalities of standard treatments for early-stage breast cancer, we addressed these gaps by examining the association between plasma cytokine levels and cognitive performance across multiple domains in women with early-stage breast cancer over two years using validated, computerized cognitive testing.


As described in previous papers (Aboalela et al. 2015, Lyon et al., 2016), women between the ages of 21–65 with stage I to IIIA breast cancer were recruited from a National Cancer Center designated Cancer Center affiliated with an urban research University in the mid-Atlantic region and multiple collaborative sites across the state. A total of 77 women with early stage (I to IIIA) breast cancer, who ranged from 23 to 71 years of age, were recruited through 5 regional cancer centers in Central Virginia. Two women withdrew from the study prior to the initial data collection yielding an N=75. To identify potential study participants, each site had a study coordinator who screened patients for eligibility. The eligibility criteria were: (1) an age of 21 years or older; (2) a diagnosis of early stage breast cancer with a scheduled clinic appointment to receive chemotherapy; and (3) female gender (males were excluded since too few male participants were available for study). Exclusion criteria were a history of: (1) a previous cancer, or chemotherapy; (2) a diagnosis of dementia; (3) active psychosis; or (4) immune-related diagnoses (e. g. multiple sclerosis; systemic lupus erythematosus). After providing informed consent (VCU IRB #HM 13194), participants were enrolled and the first study visit was scheduled prior to the initiation of chemotherapy. The five time points for evaluation in this study were: prior to the start of chemotherapy (T1), at the midpoint of chemotherapy (T2), 6 months after the initial chemotherapy (T3), one year after the initial chemotherapy (T4) and two years after the initial chemotherapy (T5). The initial assessment (T1) was conducted after surgery but prior to commencing chemotherapy in women receiving adjuvant therapy. Women receiving neoadjuvant therapy had chemotherapy prior to surgical resection. After obtaining informed consent, participants were asked to complete questionnaires and performance-based cognitive testing via a computerized system. Participants were given incentives (a $25 gift card to a local store that has both food and personal items for sale) at each data collection point over the two-year study.



Whole blood was collected at each study visit. Plasma was separated by centrifugation, and all specimens were aliquoted immediately, frozen, and stored in a −80°C freezer. A standard capture sandwich assay was used to determine the levels of different cytokines. Each captured antibody was coupled to a different bead set (Bio-Rad Laboratories, Hercules, California, USA). The system uses a liquid suspension array of 17 sets of 5.5-μm beads (Bio-Plex Human Cytokine 17-plex panel) internally dyed with different ratios of two spectrally distinct fluorochromes to assign a unique spectral address. Each set of beads was combined with a monoclonal antibody raised against interleukin (IL)-1β, IL-2, IL-4, IL-5, IL-6, IL-7, IL-8, IL-10, IL-12 (p70), IL-13, IL-17, granulocyte-colony stimulating factor (G-CSF), granulocyte-macrophage colony stimulating factor (GM-CSF), interferon gamma (IFNγγ), MCP-1 (monocyte chemotactic and activating factor), macrophage inflammatory protein (MIP-1β) or tumor necrosis factor (TNF-α). Beads were incubated first (30 min to 2 h, at room temperature) with diluted standards (serial dilutions from 1.95 to 32 000 pg/ml) and then with biotinylated detector antibodies (30 min, at room temperature). They were washed twice in phosphate-buffered saline, and incubated for 30 min at room temperature with phycoerythrin-conjugated streptavidin. Each measurement was taken in duplicate. Standard curves were generated by using the reference cytokine concentrations supplied by the manufacturer. Raw data (mean fluorescent intensity) were analyzed by Bio-Plex Manager Software (Bio-Rad Laboratories) to obtain concentration values. The lower limit of detection was <10 pg/ml (based on detectable signal >2 standard deviations (SD) above background). Samples were batch processed using plates from the same lots to reduce measurement variability.

Cognitive function

A performance-based computerized neurocognitive testing system, CNS Vital Signs (CNSVS, (Gualtieri and Johnson, 2006) was used to measure multiple cognitive domains. CNVS was standardized with a normative sample of 1,069 subjects ranging in age from 7 to 90, drawn from the American population and has been used in multiple samples in research and clinical settings (Crawford and Jonassaint, 2015, Gualtieri and Johnson, 2006). Test results are presented in subject (raw) scores, age-matched standard scores, and percentile ranks. CNSVS standard scores have a mean of 100 and a standard deviation of 15; higher scores indicate better performance on memory and attention. The CNSVS scores individual tests and calculates a report of the clinical domains of neurocognitive functioning. The clinical domains include a neurocognitive index which is an average of five neurocognitive domains including: 1) memory, 2) psychomotor speed, 3) reaction time, 4) complex attention, and 5) cognitive flexibility. There are seven subtests that yield the five clinical domains and the combined neurocognitive index. The subscales of the CNSVS have good test-retest reliability: attention (r = 0.65), memory (r = 0.66), psychomotor speed (r = 0.88), cognitive flexibility (r = 0.71), and reaction time (r = 0.75) and have been used previously in women with breast cancer (Collins et al., 2015, Ercoli et al., 2015).


Demographic, cancer and cancer treatment-related variables were collected by medical record review and participant interview. Demographic variables included: age, race, educational level, and body mass index. Body mass index (BMI) was calculated as weight divided by height squared. Cancer-related variables included breast cancer stage (TNM) and hormone receptor status. Treatment-related variables included surgery type, treatment regimen (adjuvant or neo-adjuvant), chemotherapeutic regimen, radiation status and hormonal agent status.

Statistical Analysis

All statistical analysis was performed in SAS version 9.4. Log-transformation was applied to the cytokine concentrations. To examine the cytokine log-concentrations changes across time, we fit linear mixed effects models with subject-specific random effects to account for the within subject correlation. Model-based F-tests were used to examine the overall temporal changes for each cytokines. To examine the relationship between cytokine log-concentrations and cognitive performance, we used backwards model selection procedure with the Akaike Information Criterion (AIC) (Akaike, 1974, Burnham and Anderson, 2002). The procedure is a linear regression model-selection algorithm that balances model fit with complexity to identify a parsimonious set of independent variables that predict a dependent variable. Demographic, cancer and cancer treatment-related variables were considered in the model selection as well.


Clinical and demographic characteristics

Table 1 provides demographic and clinical characteristics of the cohort (Insert Table 1 here). All participants were adult females with a recent diagnosis of early-stage breast cancer between the ages of 23 and 71. All were scheduled for chemotherapy prior to study recruitment and were available for baseline evaluation prior the onset of their chemotherapy. The participants tended to be overweight based on their body mass index (mean BMI = 29.85 +7.47). Tumor stage and grade varied (Stage: I- 27%, IIA-41%, IIB-21%, IIIA-11%; Grade: 1–7%, 2–37%, 3–56%). The majority of women received TAC docetaxel (Taxotere) doxorubicin (Adriamycin), cyclophosphamide (Cytoxan) chemotherapy regimen. A majority (79%) had radiation therapy following chemotherapy and 44% were receiving hormonal therapies by T5.

Table 1
Sample demographics and treatment characteristics

Cytokine concentrations across time

Plasma concentrations of six of the 17 cytokines and chemokines (IL-6, IL-12, IL-17, G-CSF, MIP-1β, MCP-1) changed significantly across visits. Figure 1 provides plots of the five cytokines in the sample across time. (Insert Figure 1 here). IL-6 and MIP-1β concentrations increased from baseline through T3 (i.e., immediately post-radiation), and subsequently to levels that were below baseline. From very low baseline levels, MCP-1 concentrations also increased during chemotherapy. MCP-1 concentrations were reduced at 12- and 24-months relative to the chemotherapy period, but were still elevated relative to baseline. IL-12 and IL-17 concentrations were most elevated at baseline and showed a consistent decline across time. G-CSF was elevated at baseline and remained so through the initial chemotherapy period T2, but then declined precipitously over subsequent time-points.

Figure 1
Cytokine changes over time.

Relationship of cytokine concentrations to cognitive function

Serum cytokine concentrations of specific cytokines were associated with cognitive performance on some, but not all domains of the CNSVS. The associations between cytokine concentrations and cognitive performance varied across time. Tables 26 detail the cytokines associated with the composite score for each cognitive domain, along with their beta coefficients.

Table 2
Time point 1 (T1)
Table 6
Time point 5 (T5)

Baseline cytokines and cognitive function (T1)

G-CSF concentrations were positively associated with cognitive flexibility (t=2.33, p<.03), executive functioning (t=2.12, p<.04), and psychomotor speed (t=2.20, p<.04), whereas IL-17 was negatively associated with psychomotor speed (t = −2.69, p<.01). GM-CSF was positively associated with reaction time (t=2.52, p<.02) at baseline. None of the other cytokines were associated with baseline performance in the domains of memory or complex attention (Insert Table 2 here).

Cytokines and cognitive function prior to mid chemotherapy (T2)

IL-7 concentrations were positively associated with composite memory (t=2.41, p<02) and visual memory (t=2.56, p<.02). IL-5 and IL-17 were positively associated with psychomotor speed (IL-5: t=2.18, p<.04; IL-17: t=4.94; p<.01), whereas IL-1β and IL-12 were negatively associated with psychomotor speed (IL-1β: t=−3.19, p<.01; IL-12: t=−3.16; p<.01). IL-4 and IL-1β were differentially associated with complex attention (IL-4: t=−2.16, p<.04; IL-1β: t=2.21, p<.04). IL-8 and IL-17 were differentially associated with cognitive flexibility (IL-8: t=−3.21; p<.01; IL-17: t=2.80, p<.01) and executive functioning (IL-8: t=−3.20; p<.01; IL-17: t=2.71, p<.01). IL-8 (t=2.07, p<.05) and IL-13 (t=2.75, p<.01) were positively associated with verbal memory whereas IL-12 (t=−2.39, p=.02) and IL-1β (t=−2.12, p<.04) were inversely associated with verbal memory (Insert Table 3 here).

Table 3
Time point 2 (T2)

Cytokines and cognitive function post-chemotherapy (T3)

Multiple cytokines were significantly associated with cognitive performance post-chemotherapy (T3), though again this varied by cognitive domain. IL-17 was positively associated with psychomotor speed (t=2.08, p<.05) while IL-1β was negatively associated with psychomotor speed (t=−2.03, p<.05). GM-CSF, IL-5, IL-7, and IL-12 were differentially associated with composite memory (GM-CSF: t =−2.38, p=.02; IL-7: t=−2.40; p=.02; IL-5: t=2.11; p<.04; IL-12: t=2.14, p<.04) at this time-point. Reaction time performance was differentially associated with G-CSF, IFN-γ, IL-7 and IL-17 concentrations (G-CSF: t=−3.55, p<.01; IFN- γ: t=− 3.48, p<.01, IL-7: t=2.50; p<.02; IL-17: t=2.67, p =.01). IL-7 and IL-10 were differentially associated with both cognitive flexibility (IL-7: t=2.99; p<.01; IL-10: t=−3.40, p<.01) and executive function (IL-7: t= 3.11; p<.01; IL-10: t=−3.68, p<.01). None of the cytokines were significantly associated with complex attention. G-CSF (t=2.08, p<.05) and IL-2 (t=3.48, p<.01) were positively associated with visual memory while IFN- γ (t=−2.76, p<.01) and IL-10 (t=−2.91, p<.01) were negatively associated with visual memory (Insert Table 4 here).

Table 4
Time point 3 (T3)

Cytokines and cognitive function one year post initial chemotherapy (T4)

Psychomotor speed was positively associated with MIP-1β (t=2.45, p<.02) and negatively associated with MCP-1 (t=−2.35, p<.03). Reaction time was positively associated with IL-7 (t=2.38, p=.02) and negatively associated with IL-5 (t=−2.05, p<.05). IL-4 was positively associated with complex attention (t=3.32, p<.01) and cognitively flexibility (t=2.59, p<.02) while IL-6 was negatively associated with these cognitive domains (t=−2.02, p<.05; t=−2.12, p<.04 respectively). IFN- γ and IL-8 were negatively associated with executive functioning (t=−2.17, p<.04; t=−2.94, p<.01 respectively) while IL-4 was positively associated with executive functioning (t=3.26, p<.01) (Insert Table 5 here).

Table 5
Time point 4 (T4)

Cytokines and cognitive function at two years post initial chemotherapy (T5)

IL-7 was inversely associated with composite memory (t=−2.75; p<.01) and visual memory (t=−2.99, p<.01). GM-CSF and IL-17 were significantly associated with psychomotor speed (GM-CSF: t =2.37, p<.03; IL-17: t=2.99, p<.01) whereas MCP-1 and TNF-α were inversely associated with psychomotor speed (MCP-1: t=− 2.98, p<.01, TNF-α: t=−2.92; p<.01). Complex attention was associated with MIP-1β (t=2.81, p<.01) and inversely associated with MCP-1 (t=−2.63, p<.02). IL-5 was inversely associated with verbal memory (t=−2.11, p<.04) (Insert Table 6 here).


This is one of the largest and longest cohort studies of serum cytokines and their relationship to performance across neurocognitive domains in women with early-stage breast cancer who were assessed prior to chemotherapy and at targeted intervals over a two-year time period. Significant associations were found between serum cytokine and cognitive performance, though these associations varied as function of time, with concentrations of specific cytokines differing from pre-chemotherapy baseline through the chemotherapy treatment phase, and at subsequent long-term follow-up. Overall, these findings indicate that serum cytokine levels varied over time, corresponding to the pre-chemotherapy phase, chemotherapy, and to the recovery following multiple treatment modalities including chemotherapy and radiotherapy. Serum concentrations of six of the 17 cytokines and chemokines (IL-6, IL-12p70, IL-17, G-CSF, MIPS-1β, MCP-1) changed significantly over time (Figure 1). The trajectory of change differed among these six cytokines. IL-6, and the chemotactic cytokines MIP-1β and MCP-1 increased from baseline through T3 (i.e., immediately post-chemotherapy), but then decreased subsequently over time to levels that were below baseline at the two-year follow-up. In contrast, IL-12 and IL-17 concentrations were very elevated at baseline, but then consistently declined across time, even during the chemotherapy treatment period. G-CSF was elevated at baseline and remained so through the initial chemotherapy period (T2) but then declined precipitously over subsequent time-points. The fact that these six cytokines showed three distinct trajectories of change suggests that they were likely responding differentially to the effects of cancer and cancer therapy (e.g. chemotherapy and radiation), and that these responses were affected by factors beyond the effects of chemotherapy alone.

Levels of IL-6 have been noted to be increased in women with breast cancer as compared to health controls and (Zhang and Adachi, 1999) (Benoy et al., 2002, Kozłowski et al., 2002). Likewise, serum levels of IL-6 (M =111.38 pg/ml) as well as IL-12 (M=1142.75 pg/ml) were significantly increased in breast cancer patients as compared to controls (Hussein et al., 2003). A recent study (Cheung, Ng, 2015a) also noted an increase in levels of IL-6 from prior to chemotherapy to six and 12 weeks after chemotherapy initiation. Likewise, serum IL-12 levels were significantly higher in patients than in controls, both without and after chemotherapy (Kovacs, 2001). Another study found IL-12 levels did not decrease after the initiation of anthracycline-based chemotherapy (Derin et al., 2007); however, the time-period for the study was over only two cycles of chemotherapy. Serum levels of IL-17 (median, 91.9 vs. 40.3 pg/ml, p<0.0001), and G-CSF (median, 61.0 vs. 29.4 pg/ml, p<0.0001) were significantly higher in breast cancer patients than in healthy volunteers and the levels of these markers were inversely correlated with disease outcomes (Kawaguchi et al., 2016, Schettler et al., 2014). In an in depth study of multiple chemokines, IL-8, MIP-1 alpha, MIP-1β, were increased in 11 breast cancer patients compared to controls (Li et al., 2016).

Cognitive performance in the current cohort was associated with higher levels of some of the prototypical pro-inflammatory cytokines (e.g. IL-6) (Patel et al., 2015) that have been reported to be associated with cognitive dysfunction in other studies of women with breast cancer. Yet, other cytokines emerged that provide potentially novel associations with cognitive performance in this cohort such as IL-17, which has not been well studied in the context of CRCI. IL-17 was a significant correlate across multiple cognitive domains (executive functioning, cognitive flexibility, psychomotor speed). In addition to IL-17, IL-7 correlated with cognitive performance across four domains (reaction time, executive functioning, cognitive flexibility, memory). IL-10 was associated with three cognitive domains (executive, cognitive flexibility, visual memory). Other cytokines were differentially associated with specific cognitive functions, but with less consistent and robust relationships.

Despite this evidence of an overall relationship between serum cytokine concentrations and reduced cognitive performance, the results indicate that these relationships are complex, as they vary both as a function of treatment stage, time, and cognitive domain. At baseline prior to chemotherapy, the relationship between the cytokines and cognitive function was largely limited to G-CSF, a cytokine that functions as a growth factor that stimulates the survival, proliferation, differentiation and function of neutrophil precursors and mature neutrophils and mobilizes hematopoietic stem cells from the bone marrow (Hollmén et al., 2016). A recent study, using a multi-stage mouse model of breast cancer, showed that granulocyte-colony stimulating factor (G-CSF) is the only hematopoietic growth factor to increase in serum during early tumor development (Casbon et al., 2015). G-CSF also has neurotropic effects in the brain, increasing neurogenesis and reducing apoptosis (Jeon et al., 2016, Schneider et al., 2005). In this sample, higher levels of G-CSF concentrations were positively associated with cognitive flexibility, executive functioning, psychomotor speed and reaction time at baseline, possibly reflecting its neurotropic properties. Although not well studied in the context of CRCI, G-CSF has been noted to have neurotrophic properties. Lower G-CSF plasma levels were found in 50 early Alzheimer’s disease (AD) patients in comparison with 50 age-matched healthy controls (Laske et al., 2009). GM-CSF levels had a significant positive association with reaction time. GM-CSF, a putative growth factor for microglia/macrophages, has been associated with protective functions in a mouse model of AD in terms of Amyloid β plaque accumulation and cognitive function (Schettler, Klaidman, 2014). Similarly to a recent study of 99 women with early stage breast cancer using multiplex cytokine technology and computerized cognitive testing (Cheung et al., 2015b), higher concentration of IL-4 was associated with better cognitive performance in our study. In that study, higher IL4 was associated with faster response in speed performance after the completion of the initial chemotherapy regimen (P = 0.022). IL-4 has a role in the regulation of brain immunity, with measureable downstream effects on spatial learning/memory and neurogenesis (Gadani et al., 2012). The proinflammatory cytokine, IL-17, had variable relationships with cognitive domains over time. While the observed associations between G-CSF, GM-CSF, and cognitive function are intriguing given their possible neuroprotective influence, recombinant human granulocyte colony-stimulating factor (r-metHuG-CSF; Filgrastim) G-CSF is a commonly prescribed adjunctive therapy in breast cancer patients for patients with neutropenia (Welte et al., 1996). Of note, G-CSF and GM-CSF which were positively associated with cognitive performance at baseline had the opposite relationship during the chemotherapy period (i.e., elevated concentrations corresponded with weaker cognitive performance). This relationship may have been influenced by the receipt of G-CSF-inducing medication Filgrastim (recombinant human granulocyte colony stimulating factor) by study participants with neutropenia. In these patients, higher levels could reflect the short-term effects of the medication instead of a stabilized level of endogenous G-CSF.

In contrast to relatively few significant relationships between cytokines and cognition at baseline, multiple cytokines were found to be associated with performance across cognitive domains at T2 and T3, time-points that corresponded with the chemotherapy phase and immediately following completion of chemotherapy. Better executive function and cognitive flexibility, and faster reaction times were associated with higher levels of IL-7 (a growth factor involved in hematopoiesis and stem cell function) and IL-10 (an anti-inflammatory cytokine). These positive associations likely reflect beneficial effects of IL-7 and IL-10. However, the role of IL-10 in cognition is unclear. IL-10 was negatively associated with a composite score of executive function and processing speed in a sample of 1312 elders (Tegeler et al., 2016) and higher levels of IL-10 result in impaired neuronal differentiation in an animal model of the sub-ventricular zone (SVZ) of the lateral ventricle (LV), the main neurogenic area in the adult murine brain (Perez-Asensio et al., 2013).

The relationship between IL-17 concentrations and cognitive performance differed from the patterns observed for other cytokines and changed in direction over time. IL-17 was negatively associated with psychomotor speed at baseline, though the relationship was relatively modest. Yet, at all subsequent time-points, IL-17 was positively associated with cognitive performance. This relationship was most pronounced during the chemotherapy period (T2), as elevated IL- 17 concentrations corresponded with better psychomotor speed, executive function, and cognitive flexibility. IL-17 is an important proinflammatory cytokine that may act as a bridge between early and late immune responses; this temporal factor may contribute to the varied relationships over time. Although little studied in CRCI, IL-17 levels have been associated with protective functions in a mouse model of Alzheimer’s disease (AD) in terms of Amyloid β (Aβ) plaque accumulation and cognitive function (Pappu et al., 2011, Sun et al., 2015).

The relationships between the cytokines and cognitive performance at the one-year (T4) and two-year (T5) are of particular interest to the underlying mission of the current study. Since chemotherapy had been completed previously at T4, and for over 18 months at T5, measures from these two time-points reflect a long-termer inflammatory status, free from active chemotherapy or radiotherapy treatments. Interestingly, at one year from chemotherapy initiation, a clearer pattern of relationships was noted for pro- and anti-inflammatory cytokines. There were commonalities but also some differences with respect to the cytokines that were associated with cognitive performance at these time points. For example, IL-7 continued to be positively associated with reaction time performance at T4, but was negatively associated with memory performance at T5. IL-17 was not associated with cognitive performance at T4, but was positively associated with psychomotor speed at T5. Other cytokines that were not implicated at earlier time points also emerged as correlates of cognitive performance at T4 and T5. At T4, MCP-1 and MIP-1β were inversely associated with psychomotor speed. IL-4 and IL-6 were positively associated with complex attention and cognitive flexibility; while higher levels of IL-4, IL-8, and IFN-γ were associated with better executive functioning. None of the cytokines were associated with memory performance at T4.

At T5 (two years after initiation of chemotherapy), there were fewer relationships noted among pro- and anti-inflammatory cytokines and cognitive domains than at one-year post chemotherapy. At two years, the most notable associations between cognitive domains were with IL-7 and growth factors: levels of IL-7 were inversely related to composite and visual memory; levels of MCP-I were inversely associated with psychomotor speed and complex attention; and, higher levels of MIP-1β, which is proinflammatory but also related to homeostasis, were positively related to complex attention. MCP-1, one of the key chemokines that regulates migration and infiltration of monocytes/macrophages, including natural killer cells, remained elevated at 2-years post-breast cancer treatment. Although this finding is preliminary and warrants further study, elevated levels of MCP-1 have been noted in aging individuals with mild cognitive impairment compared to healthy controls (Galimberti, Fenoglio, 2006). TNF-α was also negatively associated with psychomotor speed at T5, whereas GM-CSF was positively associated with processing speed as was MIP-Iβ. IL-6, IL-8, and IFN-y were inversely associated with multiple cognitive domains including reaction time, complex attention, cognitive flexibility, and executive function while cytokines Il-4, Il-5 and IL were positively related to reaction time complex attention, cognitive flexibility executive function.

It needs to be emphasized that the observed relationships between specific cytokine concentrations and performance in cognitive domains at one- and two- year follow-up occurred in the context of two broader longitudinal trends in the data. Most importantly, cognitive performance improved over time overall for women in the cohort. Therefore, relationships between cytokines and cognition at these points in recovery do not point to a sustained, deleterious effect of cancer treatments and decreased cognitive function in the cohort.


In this longitudinal study of women receiving chemotherapy for early-stage breast cancer, we observed heightened levels of proinflammatory cytokines IL-12, IL-17 and MIP-1β, and G-CSF before chemotherapy. These cytokine levels fluctuated over the breast cancer treatment period and dropped precipitously at two-year post-breast cancer treatment. The observed associations between cytokines and cognitive performance provide evidence that not only prototypical cytokines (i.e. IL-6, TNF-α, and IL1-β) but also cytokines from multiple classes may contribute to the inflammatory environment that is associated with cognitive dysfunction. As many of these cytokines have been rarely assessed in prior studies, further research is warranted to determine their significance to contributing to cognitive performance in women undergoing treatment for early- stage breast cancer. Given the changes in cytokine levels over time and the relationships among cytokines and multiple domains of cognitive functioning, further study of these relationships is needed to confirm these relationships within the context of breast cancer treatment and survivorship.


  • Levels of proinflammatory cytokines IL-12, IL-17 and MIP-1β, as well as G-CSF were elevated prior to chemotherapy.
  • Levels of six cytokines (IL-6, IL-12, IL-17, G-CSF, MIPS-1β, and MCP-1) varied markedly over time.
  • Cognitive outcomes in multiple domains were associated with cytokine levels with notable changes over time.


Author declaration

All authors have seen and approved the final version of the manuscript being submitted. The article is the authors’ original work, hasn’t received prior publication and isn’t under consideration for publication elsewhere.

Conflict of Interest

The authors declare that they have no conflict of interest.

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

Contributor Information

Debra E. Lyon, University of Florida, College of Nursing, 804 895-0055.

Ronald Cohen, Center for Cognitive Aging and Memory, University of Florida, Institute on Aging.

Huaihou Chen, College of Public Health & Health Professions, College of Medicine, University of Florida.

Debra L. Kelly, University of Florida, College of Nursing.

Nancy L. McCain, Adult Health and Nursing Systems, Virginia Commonwealth University.

Angela Starkweather, Center for Advancement in Managing Pain, University of Connecticut, School of Nursing.

Jamie Sturgill, Biobehavioral Laboratory Services, Department of Family and Community Health Nursing, Virginia Commonwealth University.

Colleen K. Jackson-Cook, Cytogenetic Diagnostics Laboratory, Virginia Commonwealth University.


  • Aboalela N, Lyon D, Elswick R, Jr, Kelly DL, Brumelle J, Bear HD, et al. Perceived Stress Levels, Chemotherapy, Radiation Treatment and Tumor Characteristics Are Associated with a Persistent Increased Frequency of Somatic Chromosomal Instability in Women Diagnosed with Breast Cancer: A One Year Longitudinal Study. PloS one. 2015;10:e0133380. [PMC free article] [PubMed]
  • Akaike H. A new look at the statistical model identification. IEEE Transactions on Automatic Control. 1974;19:716–23.
  • Aluise CD, Sultana R, Tangpong J, Vore M, Clair DS, Moscow JA, et al. Chemo brain (chemo fog) as a potential side effect of doxorubicin administration: role of cytokine-induced, oxidative/nitrosative stress in cognitive dysfunction. Chemo Fog: Springer; 2010. pp. 147–56. [PubMed]
  • Balkwill F, Charles KA, Mantovani A. Smoldering and polarized inflammation in the initiation and promotion of malignant disease. Cancer cell. 2005;7:211–7. [PubMed]
  • Benoy I, Salgado R, Colpaert C, Weytjens R, Vermeulen PB, Dirix LY. Serum interleukin 6, plasma VEGF, serum VEGF, and VEGF platelet load in breast cancer patients. Clinical breast cancer. 2002;2:311–5. [PubMed]
  • Bentzen SM. Preventing or reducing late side effects of radiation therapy: radiobiology meets molecular pathology. Nature Reviews Cancer. 2006;6:702–13. [PubMed]
  • Briones TL, Woods J. Dysregulation in myelination mediated by persistent neuroinflammation: possible mechanisms in chemotherapy-related cognitive impairment. Brain, behavior, and immunity. 2014;35:23–32. [PMC free article] [PubMed]
  • Burnham KP, Anderson DR. Model Selection and Multimodel Inference: A Practical-Theoretic Approach. 2. New York: Springer-Verlag; 2002.
  • Casbon A-J, Reynaud D, Park C, Khuc E, Gan DD, Schepers K, et al. Invasive breast cancer reprograms early myeloid differentiation in the bone marrow to generate immunosuppressive neutrophils. Proceedings of the National Academy of Sciences. 2015;112:E566–E75. [PubMed]
  • Cheung YT, Ng T, Shwe M, Ho HK, Foo KM, Cham MT, et al. Association of proinflammatory cytokines and chemotherapy-associated cognitive impairment in breast cancer patients: a multi-centered, prospective, cohort study. Annals of Oncology. 2015a;26:1446–51. [PMC free article] [PubMed]
  • Cheung YT, Ng T, Shwe M, Ho HK, Foo KM, Cham MT, et al. Association of proinflammatory cytokines and chemotherapy-associated cognitive impairment in breast cancer patients: a multi-centered, prospective, cohort study. Annals of Oncology. 2015b [PMC free article] [PubMed]
  • Cohen RA, de la Monte S, Gongvatana A, Ombao H, Gonzalez B, Devlin KN, et al. Plasma cytokine concentrations associated with HIV/hepatitis C coinfection are related to attention, executive and psychomotor functioning. Journal of neuroimmunology. 2011;233:204–10. [PMC free article] [PubMed]
  • Collins B, Mackenzie J, Stewart A, Bielajew C, Verma S. Cognitive effects of hormonal therapy in early stage breast cancer patients: a prospective study. Psycho-Oncology. 2009;18:811–21. [PubMed]
  • Collins B, Paquet L, Dominelli R, White A, MacKenzie J. Metamemory function in chemotherapy-treated patients with breast cancer: an explanation for the dissociation between subjective and objective memory measures? Psycho-Oncology. 2015 [PubMed]
  • Crawford RD, Jonassaint CR. Adults with sickle cell disease may perform cognitive tests as well as controls when processing speed is taken into account: a preliminary case–control study. Journal of advanced nursing. 2015 [PMC free article] [PubMed]
  • Dantzer R, O’Connor JC, Freund GG, Johnson RW, Kelley KW. From inflammation to sickness and depression: when the immune system subjugates the brain. Nature reviews neuroscience. 2008;9:46–56. [PMC free article] [PubMed]
  • Derin D, Soydinc HO, Guney N, Tas F, Camlica H, Duranyildiz D, et al. Serum IL-8 and IL-12 levels in breast cancer. Medical Oncology. 2007;24:163–8. [PubMed]
  • Dethlefsen C, Højfeldt G, Hojman P. The role of intratumoral and systemic IL-6 in breast cancer. Breast Cancer Research and Treatment. 2013;138:657–64. [PubMed]
  • Dursun E, Gezen-Ak D, Hanağası H, Bilgiç B, Lohmann E, Ertan S, et al. The interleukin 1 alpha, interleukin 1 beta, interleukin 6 and alpha-2-macroglobulin serum levels in patients with early or late onset Alzheimer’s disease, mild cognitive impairment or Parkinson’s disease. Journal of neuroimmunology. 2015;283:50–7. [PubMed]
  • Ercoli L, Petersen L, Hunter A, Castellon S, Kwan L, Kahn-Mills B, et al. Cognitive rehabilitation group intervention for breast cancer survivors: results of a randomized clinical trial. Psycho-Oncology. 2015;24:1360–7. [PubMed]
  • Gadani SP, Cronk JC, Norris GT, Kipnis J. IL-4 in the brain: a cytokine to remember. The Journal of Immunology. 2012;189:4213–9. [PMC free article] [PubMed]
  • Galimberti D, Fenoglio C, Lovati C, Venturelli E, Guidi I, Corrà B, et al. Serum MCP-1 levels are increased in mild cognitive impairment and mild Alzheimer’s disease. Neurobiology of aging. 2006;27:1763–8. [PubMed]
  • Gualtieri CT, Johnson LG. Reliability and validity of a computerized neurocognitive test battery, CNS Vital Signs. Arch Clin Neuropsychol. 2006;21:623–43. [PubMed]
  • Hartman SJ, Marinac CR, Natarajan L, Patterson RE. Lifestyle factors associated with cognitive functioning in breast cancer survivors. Psycho-Oncology. 2015;24:669–75. [PMC free article] [PubMed]
  • Heesen C, Schulz K, Fiehler J, Von der Mark U, Otte C, Jung R, et al. Correlates of cognitive dysfunction in multiple sclerosis. Brain, behavior, and immunity. 2010;24:1148–55. [PubMed]
  • Hollmén M, Karaman S, Schwager S, Lisibach A, Christiansen AJ, Maksimow M, et al. G-CSF regulates macrophage phenotype and associates with poor overall survival in human triple-negative breast cancer. Oncoimmunology. 2016;5:e1115177. [PMC free article] [PubMed]
  • Hurria A, Rosen C, Hudis C, Zuckerman E, Panageas KS, Lachs MS, et al. Cognitive function of older patients receiving adjuvant chemotherapy for breast cancer: a pilot prospective longitudinal study. Journal of the American Geriatrics Society. 2006;54:925–31. [PubMed]
  • Hussein MZ, Al Fikky A, Abdel BI, Attia O. Serum IL-6 and IL-12 levels in breast cancer patients. The Egyptian journal of immunology/Egyptian Association of Immunologists. 2003;11:165–70. [PubMed]
  • Janelsins MC, Kesler SR, Ahles TA, Morrow GR. PREVALENCE, MECHANISMS, AND MANAGEMENT OF CANCER-RELATED COGNITIVE IMPAIRMENT. International review of psychiatry (Abingdon, England) 2014;26:102–13. [PMC free article] [PubMed]
  • Janelsins MC, Kohli S, Mohile SG, Usuki K, Ahles TA, Morrow GR. Seminars in oncology. Elsevier; 2011. An update on cancer-and chemotherapy-related cognitive dysfunction: current status; pp. 431–8. [PMC free article] [PubMed]
  • Jeon SG, Kim KA, Chung H, Choi J, Song EJ, Han S-Y, et al. Impaired Memory in OT-II Transgenic Mice Is Associated with Decreased Adult Hippocampal Neurogenesis Possibly Induced by Alteration in Th2 Cytokine Levels. Molecules and Cells. 2016;39:603. [PMC free article] [PubMed]
  • Kawaguchi K, Suzuki E, Yamamoto Y, Saito K, Toi M. Abstract P4-04-22: Upregulation of the interleukin 17/granulocyte-colony stimulating factor/fibroblast growth factor axis in breast cancer and its negative association with treatment outcome. Cancer Research. 2016;76:P4-04-22.
  • Kovacs E. The serum levels of IL-12 and IL-16 in cancer patients. Relation to the tumour stage and previous therapy. Biomedicine & pharmacotherapy. 2001;55:111–6. [PubMed]
  • Kozłowski L, Zakrzewska I, Tokajuk P, Wojtukiewicz M. Concentration of interleukin-6 (IL-6), interleukin-8 (IL-8) and interleukin-10 (IL-10) in blood serum of breast cancer patients. Roczniki Akademii Medycznej w Bialymstoku (1995) 2002;48:82–4. [PubMed]
  • Laske C, Stellos K, Stransky E, Leyhe T, Gawaz M. Decreased plasma levels of granulocyte-colony stimulating factor (G-CSF) in patients with early Alzheimer’s disease. Journal of Alzheimer’s Disease. 2009;17:115–23. [PubMed]
  • Lee PE, Tierney MC, Wu W, Pritchard KI, Rochon PA. Endocrine treatment-associated cognitive impairment in breast cancer survivors: evidence from published studies. Breast cancer research and treatment. 2016;158:407–20. [PubMed]
  • Li L, Chen L, Zhang W, Liao Y, Chen J, Shi Y, et al. Serum cytokine profile in patients with breast cancer. Cytokine. 2016
  • Lyon DE, Cohen R, Chen H, Kelly DL, Starkweather A, Ahn H-C, et al. The relationship of cognitive performance to concurrent symptoms, cancer-and cancer-treatment-related variables in women with early-stage breast cancer: a 2-year longitudinal study. Journal of cancer research and clinical oncology. 2016:1–14. [PMC free article] [PubMed]
  • Mandelblatt JS, Clapp JD, Luta G, Faul LA, Tallarico MD, McClendon TD, et al. Long-term trajectories of self-reported cognitive function in a cohort of older survivors of breast cancer: CALGB 369901 (Alliance) Cancer. 2016 n/a-n/a. [PMC free article] [PubMed]
  • Miura K, Ando S, Imai T. The association of cognitive fatigue with menopause, depressive symptoms, and quality of life in ambulatory breast cancer patients. Breast Cancer. 2016;23:407–14. [PubMed]
  • Pappu R, Ramirez-Carrozzi V, Sambandam A. The interleukin-17 cytokine family: critical players in host defence and inflammatory diseases. Immunology. 2011;134:8–16. [PubMed]
  • Pasic J, Levy WC, Sullivan MD. Cytokines in depression and heart failure. Psychosomatic Medicine. 2003;65:181–93. [PubMed]
  • Patel SK, Wong AL, Wong FL, Breen EC, Hurria A, Smith M, et al. Inflammatory Biomarkers, Comorbidity, and Neurocognition in Women With Newly Diagnosed Breast Cancer. Journal of the National Cancer Institute. 2015:107. [PMC free article] [PubMed]
  • Perez-Asensio FJ, Perpiñá U, Planas AM, Pozas E. Interleukin-10 regulates progenitor differentiation and modulates neurogenesis in adult brain. J Cell Sci. 2013;126:4208–19. [PubMed]
  • Schettler S, Klaidman L, Vinters H, Stark E. Examining the immunomodulatory effects of GM-CSF and IL-12 in a mouse model of Alzheimer’s disease (728.2) The FASEB Journal. 2014:28.
  • Schneider A, Kuhn H-G, Schäbitz W-R. Review A Role for G-CSF (Granulocyte-Colony Stimulating Factor) in the Central Nervous System. Cell Cycle. 2005;4:1753–7. [PubMed]
  • Sun J, Zhang S, Zhang X, Zhang X, Dong H, Qian Y. IL-17A is implicated in lipopolysaccharide-induced neuroinflammation and cognitive impairment in aged rats via microglial activation. Journal of neuroinflammation. 2015;12:165. [PMC free article] [PubMed]
  • Swardfager W, Lanctôt K, Rothenburg L, Wong A, Cappell J, Herrmann N. A meta-analysis of cytokines in Alzheimer’s disease. Biological psychiatry. 2010;68:930–41. [PubMed]
  • Tegeler C, O’Sullivan JL, Bucholtz N, Goldeck D, Pawelec G, Steinhagen-Thiessen E, et al. The inflammatory markers CRP, IL-6, and IL-10 are associated with cognitive function—data from the Berlin Aging Study II. Neurobiology of aging. 2016;38:112–7. [PubMed]
  • Welte K, Gabrilove J, Bronchud M, Platzer E, Morstyn G. Filgrastim (r-metHuG-CSF): the first 10 years. Blood. 1996;88:1907–29. [PubMed]
  • Wilson CJ, Finch CE, Cohen HJ. Cytokines and cognition—the case for a head-to-toe inflammatory paradigm. Journal of the American Geriatrics Society. 2002;50:2041–56. [PubMed]
  • Zhang G, Adachi I. Serum interleukin-6 levels correlate to tumor progression and prognosis in metastatic breast carcinoma. Anticancer research. 1999;19:1427–32. [PubMed]