Cross-sectional studies suggest that clinical insomnia is associated with immune downregulation. However, there is a definite need for experimental studies on this question. The goal of this randomized controlled study was to assess the effect of an 8-week cognitive-behavioral therapy (CBT) for chronic insomnia on immune functioning of breast cancer survivors. Previous analyses of this study showed that CBT was associated with improved sleep and quality of life, and reduced psychological distress.

Fifty-seven women with chronic insomnia secondary to breast cancer were randomly assigned to CBT (n = 27) or to a waiting-list control condition (WLC; n = 30). Peripheral-blood samples were taken at baseline and post-treatment (and postwaiting for WLC patients), as well as at 3-, 6-, and 12-month follow-up for immune measures, including enumeration of blood cell counts (ie, WBCs, monocytes, lymphocytes, CD3+, CD4+, CD8+, and CD16+/CD56+) and cytokine production (ie, interleukin-1-beta [IL-1β] and interferon gamma [IFN-γ]).

Patients treated with CBT had higher secretion of IFN-γ and lower increase of lymphocytes at post-treatment compared with control patients. Pooled data from both treated groups indicated significantly increased levels of IFN-γ and IL-1β from pre- to post-treatment. In addition, significant changes in WBCs, lymphocytes, and IFN-γ were found at follow-up compared with post-treatment.

This study provides some support to the hypothesis of a causal relationship between clinical insomnia and immune functioning. Future studies are needed to investigate the clinical impact of such immune alterations.

The psychoneuroimmunologic (PNI) model of cancer, though not yet validated, proposes that psychological difficulties can influence cancer progression through alterations of immune functioning.1-3 In the last decade, PNI research has flourished and a large range of psychological variables, including stress and depression, have been associated with changes in immune functioning.4-6

Sleep difficulties affect between 30% and 50% of cancer patients7 and are commonly associated with psychological disturbances such as depression and anxiety.8 Sleep has typically been considered as a confounding factor in PNI research (ie, a factor that should be controlled for), because of its potential impact on immune functioning. More recently, relationships of immune function with sleep deprivation and sleep-wake rhythms have started to be investigated on their own.9

The effect of experimental sleep deprivation on the immune system has been the focus of several studies. Despite some inconsistencies, the findings generally suggest that acute and total sleep deprivation (ie, 48 to 72 hours without sleep) is associated with immune activation,10-13 whereas partial sleep deprivation (ie, for a portion of the night) is associated with immune downregulation.14-16

Although these data provide evidence of the interconnection between sleep and immune function, they may not generalize to clinical insomnia. Indeed, experimental sleep deprivation and clinical insomnia are two highly distinct phenomena; sleep deprivation shortens sleep because of reduced opportunity to sleep, whereas clinical (or naturalistic) insomnia is a complaint of reduced sleep despite having the opportunity to sleep. Furthermore, the immune changes observed after experimental sleep deprivation are transient and return to basal levels after a night of recovery sleep,17 a phenomenon that may not generalize to patients with clinical insomnia who have repeated nights of poor sleep during several weeks or months.

Some cross-sectional studies conducted in the context of depression have suggested a relationship between clinical sleep difficulties and immune downregulation, as indicated by reduced natural killer (NK) cell activity or reduced levels of some lymphocyte subpopulations in the circulating blood.18-20 More recently, our research team showed that otherwise healthy individuals with chronic primary insomnia had lower levels of some lymphocyte subpopulations (ie, CD3+, CD4+, CD8+, and CD16+/CD56+) compared with good sleepers.21 However, no difference in NK cell activity, interleukin-1-beta (IL-1β), IL-2, and interferon gamma (IFN-γ) production was observed between the two groups. Another research group showed that primary insomnia was associated with decreased NK cell activity compared with controls, but no between-group differences were observed on IL-2 production, IL-2–stimulated NK cell activity, or lymphocyte subsets (CD3+, CD16+/CD56+ cells).22 Finally, a study using serial 24-hour plasma measures showed no significant difference between young patients with chronic insomnia and matched controls on the mean IL-6 and tumor necrosis factor (TNF) secretions, but found a daytime shift of IL-6 and TNF secretion in patients with insomnia.23

To examine more precisely the relationship between clinical insomnia and immune alterations, experimental studies are needed. Clinical studies assessing the effect of insomnia treatment on immune function constitute an advantageous alternative. If insomnia has a deleterious effect on immune function, a successful treatment of insomnia should have the opposite effect on immunity. Furthermore, because it is difficult, and even undesirable, to increase immune functioning above a certain threshold (autoimmune disorders are associated with overly active immunity), ideally this should be studied in patients with compromised immunity,24 such as women who have been treated for breast cancer. Previous studies generally have found that psychological interventions can improve immunologic functioning of breast cancer patients, including increased lymphocyte numbers, NK cell activity, lymphocyte proliferation, and IFN-γ production.25-29 This suggests that breast cancer patients constitute an appropriate group for conducting these types of studies.

The main goal of this study was to assess the impact of a psychological treatment (ie, cognitive-behavioral therapy [CBT]) for insomnia on immune functioning (ie, enumerative measures of lymphocyte subpopulations, NK cell activity, cytokine production) among women who had been treated for breast cancer. In the companion study,30 CBT for insomnia secondary to cancer was associated with significant reductions of sleep difficulties, as well as improvement of mood and overall quality of life. A secondary goal of the present study was to assess to what extent the impact of insomnia treatment on immune functioning was mediated by actual sleep improvements, and changes in psychological distress, and fatigue levels.

Participants

The participant selection procedure and characteristics are fully described in the companion report.30 Briefly, participants were 57 women meeting criteria for an insomnia syndrome (mean age, 54.1 years; standard deviation [SD], 7.4 years). In all patients, insomnia was chronic (ie, duration of 6 months or more) and was judged to be secondary to breast cancer (ie, caused or significantly aggravated by the breast cancer diagnosis or treatment). Most patients initially received a diagnosis of stage I or II breast cancer (94.7%). On average, women had received their cancer diagnosis 41.9 months ago (SD, 49.4 months). The mean duration of insomnia was 10.1 year (SD, 10.5 years) and 45.6% of the patients stated that their insomnia was caused by cancer, whereas the remaining patients (54.4%) reported that their pre-existing insomnia was significantly aggravated with cancer.

Experimental Design

Participants were randomly assigned to one of the two following conditions: CBT or waiting-list control condition (WLC). Because the recruitment was somewhat slow and treatment was conducted in small groups, the random allocation of patients was done by blocks of five to six patients. Participants assigned to the control group waited a minimum of 8 weeks, which corresponded to the duration of CBT, were assessed again on outcome variables, including completion of a sleep diary for a 2-week period, and then received CBT. Additional evaluations were conducted 3, 6, and 12 months after the end of their respective treatment to assess the maintenance of treatment effects over time. The study was approved by the ethical review boards of L'Hôtel-Dieu de Québec and Université Laval (Québec, Canada).

Procedure

Study procedures are fully described in the companion report.30 Hereafter, we report only procedures relevant to this study's objectives. All participants completed a battery of self-report scales as part of an information session about the study. This included the Insomnia Severity Index (ISI),31 the Hospital Anxiety and Depression Scale (HADS),32,33 the Multidimensional Fatigue Inventory (MFI),34,35 and the Health Behaviors Questionnaire (HBQ). The HBQ was elaborated by our research team to evaluate health behaviors that may potentially confound the relationship between insomnia and immune functioning.21 Then, the first blood samples for immune measures were taken by a registered nurse. To control for diurnal variations, all blood samples at this and other study assessments were drawn at the same time (± 1 hour) in early evening. In addition, participants were asked to complete a daily sleep diary for a period of 2 weeks. Patients assigned to the treatment condition received the insomnia treatment immediately, whereas those assigned to WLC waited 8 weeks and were re-evaluated on the same variables (ie, sleep, health behaviors, immune measures) before receiving the insomnia treatment. The same assessments were repeated at post-treatment, as well as at 3-, 6-, and 12-month follow-up.

The insomnia treatment consisted of eight weekly sessions of approximately 90 minutes, offered in groups of four to six patients, and based on clinical procedures described in a treatment manual.36 This multimodal approach combined behavioral (ie, stimulus control and sleep restriction strategies), cognitive (ie, cognitive restructuring), and educational (ie, sleep hygiene) strategies. Strategies to cope with fatigue and stress were also integrated in the intervention. An optional booster session was offered to participants 1 month after the end of the treatment. The intervention content is more fully described in Quesnel et al.37

Immunologic Measures

At each assessment point, 36 mL of venous blood was collected in four heparinized tubes. The tubes were centrifuged at 1,800 rpm for 5 minutes at room temperature and the buffy coat was collected (3 mL by tube). Analyses were performed by laboratory personnel blinded to the patients' random assignment.

WBCs subsets were determined in the whole blood by three-color direct immunofluorescence using a flow cytometer (Coulter EPICS Elite ESP; Beckman Coulter, Miami, FL). A minimum of 10,000 cells per sample were analyzed. To analyze lymphocyte surface antigens, monoclonal antibodies (mAbs) were directly conjugated with fluorescein isothiocyanate (FITC), phycoerythrin (PE), or peridinin chlorophyll protein (PerCP). Briefly, for each subset analysis, 10 μL of mAb (TriTEST) was added to 50 μL buffy coat and incubated for 20 minutes. Erythrocytes were then disintegrated using OPTILYSE-C reagent (Immunotech, Miami, FL). Enumeration by flow cytometry included the following cells: T cells (CD3+; CD3 FITC), T-helper cells (CD3+CD4+; CD3 FITC/CD4 PE/CD45 PerCP), T-suppressor/cytotoxic cells (CD3+CD8+; CD3 FITC/CD8 PE/CD45 PerCP), and NK cells (CD3/CD16+CD56+; CD3 FITC/CD16+CD56 PE/CD45 PerCP). All mAbs and immunofluorescence reagents were purchased from Becton Dickinson (San Jose, CA). The absolute number per unit volume bearing each lymphocyte marker was determined by multiplying data obtained by flow cytometry with the absolute lymphocyte count derived from the CBC and differential. The CBC was performed within 2 hours after the blood was drawn, whereas flow cytometry was conducted the next morning (within 12 to 15 hours after the blood was drawn).

Peripheral-blood lymphocytes were separated by density gradient centrifugation on a Ficoll-Hypaque gradient (Amersham Pharmacia Biotech, Piscataway, NJ) the morning after the blood was drawn (within 12 to 15 hours) and were stored (20% dimethyl sulfoxide; 80% fetal bovine serum) at −80°C until the assay was performed within 3 years after. The NK cytotoxicity was determined by flow cytometry.38 This method is fast, reliable, and correlates well with the standard 51CR-release assay while avoiding the use of radioactive material.39-41 Peripheral-blood lymphocytes were thawed rapidly in a 37°C water bath, washed with RPMI-1640 (Cellgro; Winset Canadian Laboratories, St Bruno, Canada) plus 10% fetal bovine serum, and kept for one night in a humidified 5% CO2 atmosphere at 37°C.

The following morning, they were transferred in culture flasks and kept for 30 minutes in a humidified 5% CO2 atmosphere at 37°C. Effector cells were then washed, counted, and adjusted to 5 × 106/mL. Target cells, K562, a human erythroleukemic cell line (CC L243; American Type Culture Collection, Rockville, MD) in log phase were washed, counted, and adjusted to 1 × 105/mL. Then, effector cells and target cells were mixed at four ratios of effector to target cells (50:1, 25:1, 12.5:1, and 6.25:1) and incubated at 37°C in a humidified 5% CO2 incubator for 10 minutes to promote conjugate formation. Then, 10 μL of propidium iodide working solution (100 μL/mL) was added into each tube and incubated for 90 minutes at 37°C in a 5% CO2 atmosphere. Finally, cells were stored in a dark ice bath and flow cytometric data acquisition was performed with a flow cytometer (Coulter EPICS Elite ESP, Beckman Coulter). In a previous study in which the same procedure was used, we found that cryopreservation of blood samples had no influence on NK cell activity.21

For determination of IL-1β and IFN-γ production, a whole-blood assay was performed.42 These parameters were chosen because there was a large body of evidence of a bidirectional relationship between these variables and sleep at the time the study was initiated,11,43 as well as some evidence of a relationship with breast cancer prognosis.44 For stimulation of IL-1β, aliquots of 50 μL of buffy coat were resuspended in 445 μL of RPMI-1640 medium and 5 μL lipopolysaccharide from Escherichia coli (0.1 mg/mL). For stimulation of IFN-γ, aliquots of 50 μL of buffy coat were resuspended in 440 μL of RPMI-1640 medium and 10 μL phytohemagglutinin (2%) was added. Every sample was stimulated in duplicate the morning after the blood was drawn (within 12 to 15 hours). Then, the samples were incubated for a minimum of 72 hours at 37°C in a 5% CO2 atmosphere. The supernatant were stored at −80°C until the assay was performed within 3 years after the blood was drawn. A minimum incubation time of 72 hours was chosen on the basis of previous kinetic studies indicating that it provides a good estimate for the production of cytokines assessed in this study.45 All cytokine levels were measured by enzyme-linked immunosorbent assay kits (Biosource International, Camarillo, CA). The sensitivities of the assays were 1 pg/mL for IL-1β and 4 pg/mL for IFN-γ.

Statistical Analyses

All data were inspected carefully to identify missing data and outliers and to assess normality.46 Descriptive and inferential statistics were completed using SAS 8.2 statistical software (SAS Institute, Cary, NC).47 For all inferential tests, α was fixed at .05 (two-tailed). Data were analyzed within an intent-to-treat framework. Linear mixed models were used to test group, time, and group-time interaction effects for all continuous dependent variables. A priori contrasts were used to break down these effects. Satterthwaite F tests were computed because they typically are more robust to non-normality, unbalanced data, and violations of multisample sphericity.48

The analyses were based on a split-plot group (two conditions) -time (five assessments: pretreatment, post-treatment, and 3-, 6-, and 12-month follow-up) randomized design. Two subsets of analyses were performed. First, analyses were conducted to determine whether treated patients had greater changes on the range of dependent variables at post-treatment compared with patients in the control group after their waiting period. These findings are reported in Group Comparisons; because of this objective, only significant group-time interactions are reported. Then, because of the nonsignificant group effects obtained on most immune variables (ie, WBC; monocytes; CD3+, CD4+, CD8+, and CD16+/CD56+ cells; and IFN-γ and IL-1β production), data of both groups were pooled together to confirm, with a larger sample size, the changes associated with the intervention at post-treatment and to evaluate whether changes observed at post-treatment were maintained at follow-up assessments. These findings are reported in Pooled Analyses and only significant time effects are reported (ie, from pre- to post-treatment and from post-treatment to follow-up). In addition, NK cell activity data were submitted to a factorial randomized block mixed-models analysis with two within-subject factors (time, concentration). To test whether activity curves (concentration effect) were significantly different between groups and across time, higher order interactions were decomposed and treatment-contrast interactions were tested.49

In accordance with the strategy suggested by Frigon and Laurencelle,50 various covariates (demographics, medical factors, and health behaviors) were tested to assess their capacity to reduce the error term. To be included in the mixed-model analyses as a covariate, a variable had to meet these two criteria: significant between-group differences and significant reduction of error variance of more than one dependent variable. In the final models for CD3+, CD4+, CD8+, lymphocytes, monocytes, and NK cells, as well as IL-1β and IFN-γ production, these covariates were education level, body mass index, time elapsed since the end of a hormonal treatment, physical activity level (in metabolic equivalents), alcohol consumption in the last 48 hours, number of meals typically eaten per day, compliance with the Canada's Food Guide to Healthy Eating recommendations,51 the usage of vitamins, and the usage of the following medications: Phenobarbital, hydrochlorothiazide, omeprazole, and fosinopril sodium. For final analyses on NK cell activity, the covariates were age, presence of another physical disease, frequency of a regular meal schedule, and usage of hydrochlorothiazide and soy protein. In addition, because NK cell activity was assessed on 43 different days, covariance associated with the assay effect was also controlled for in the mixed models.

Finally, additional analyses were conducted on the pooled data set to test for the presence of a mediator role of changes in sleep, psychological distress, and fatigue indices in the effect of the intervention (time effect) on immune measures. The same demographics, medical variables, and health behaviors identified were controlled in those analyses. Because we were interested in finding potential mechanisms through which the treatment of insomnia could lead to immune changes, only immune parameters on which significant time effects were obtained were analyzed for mediating effects. In accordance with the two-step methodology proposed by Baron and Kenny,52 linear mixed-model analyses were first conducted to obtain the total time effect. The direct time effect was obtained after the introduction of sleep (ie, total sleep time and sleep efficiency indices, based on the daily sleep diary; and the ISI score), psychological distress (ie, total HADS score), and fatigue (ie, total MFI score) variables were introduced as covariates. Then, indirect time effects for each potential mediator were computed as the product of the covariate parameter (β) and the time effect on the covariate (α).53 The z′ statistic was finally computed and compared with the table in the appendix of MacKinnon et al53 (ie, null hypothesis = no mediating effect; n = 100) to determine the statistical significance of the mediating effect.

Group Comparisons

After controlling for the covariates, mixed-model analyses revealed significant group-time interactions on two immune variables. Specifically, patients treated with CBT had a higher increase in IFN-γ secretion (F1,36 = 6.75; P < .01), and a lower increase of lymphocyte count (F1,50 = 4.46; P < .05) at post-treatment compared with control patients after their waiting period (Table 1; Figs 1 and 2). Although data illustrated in Figure 1 suggest a higher increase in IL-1β production in the treated group, the group-time interaction was not significant (F1,51 = 1.01; P = .32). In addition, the analyses revealed no significant group-time-concentration interaction on NK cell activity after controlling for the covariates (F3,126 = 0.85; P = .47; Table 1; Fig 3).

Pooled Data

Mixed-model analyses conducted on pooled data from both groups (after WLC control patients received CBT) revealed significant time effects on several immune variables, after controlling for the covariates. Specifically, increased secretion of IFN-γ (F1,130 = 6.61; P < .01) and IL-1β (F1,148 = 4.01; P < .05) from pre- to post-treatment were found. In addition, increased levels of WBC (F3,149 = 4.50; P < .01) and lymphocytes (F3,146 = 4.16; P < .01) were found at follow-up compared with post-treatment. With regard to IFN-γ secretion, a significant change from post-treatment to follow-up was found (F3,146 = 4.16; P < .01) but this change was in both directions depending on the time assessment (Table 1; Figs 1 and 2). Finally, the mixed-model analyses revealed no significant time-concentration interaction on NK cell activity after controlling for the covariates (F12,659 = 1.43; P = .15), despite a tendency for increased activity at 3- and 6-month follow-up (ratio 25:1) and decreased activity at 12-month follow-up (ratio 50:1) compared with pretreatment (Table 1; Fig 3).

Mediating Factors

Analyses were conducted using the pooled data set to determine to what extent changes in sleep, psychological distress, and fatigue indices mediated the significant changes obtained on immune functioning at post-treatment (on IFN-γ and IL-1β secretion) and at follow-up (on WBC and lymphocyte counts). For the latter variables, data of the 3-month follow-up were used because data illustrated in Figure 2 indicated that WBC and lymphocytes first increased at that time. The results indicated a small but significant mediating effect of reduced psychological distress in the effect of CBT on increased IFN-γ secretion at post-treatment, which explained 18% of the absolute change (Table 2). In addition, large and significant mediating effects of decreased insomnia scores were obtained in the effect of CBT on the absolute count of WBC and lymphocytes at the 3-month follow-up, which explained 47% and 55% of the absolute change, respectively. Finally, a small but significant mediating effect of reduced psychological distress scores was obtained for WBC at 3-month follow-up, which explained 13% of the absolute change. It is noteworthy that significant mediating effects of changes in fatigue scores were obtained on IFN-γ, IL-1β, and WBC levels, but in the opposite direction with reduced fatigue being associated with reduced immunity.

The main goal of this study was to assess the effect of a psychological treatment for insomnia on immune functioning of breast cancer survivors. The results provide some support for the hypothesis that treating chronic insomnia can alter immune functioning among this population. These study findings are consistent with the previous literature showing positive effects of a psychological intervention on some immune variables of breast cancer patients.25-29 This is also consistent with the increasing empirical literature of a bilateral communication between sleep and cytokine expression, with both evidence for an effect of experimental sleep deprivation on cytokines secretion and for an effect of cytokines administration on sleep and sleep depth.17,54,55

Overall, evidence for an immune enhancement associated with the insomnia treatment was more convincing for cytokines. In fact, only IFN-γ secretion was increased to a significantly greater degree in the treated group at post-treatment compared with the control group after their waiting period, and only IFN-γ and IL-1β production were found to be significantly enhanced from pre- to post-treatment when the pooled data set was used. The effects remained the same when the analyses were repeated to control for the number of lymphocytes and monocytes, thus suggesting that these changes were functional rather than simply reflecting a redeployment of cells. However, only increased IL-1β secretion was found to be maintained during the 1-year follow-up phase, as indicated by an absence of a significant time effect from post-treatment to follow-up on this variable, whereas changes in both directions were found for IFN-γ secretion during that period. Figure 1 also reveals a great deal of variability across groups with regard to the evolution of these immune changes during the follow-up period.

Findings on other immune variables were more equivocal. The fact that two immune variables (ie, WBC and lymphocytes) increased significantly only from post-treatment to the follow-up phase (analyses on the pooled data set), rather than during treatment, is difficult to attribute with certainty to the insomnia treatment, although a delayed effect is plausible. In fact, there is some evidence of a deferred effect of a psychological intervention on immune functioning of cancer patients.26,56 This study revealed no significant effect of CBT for insomnia on NK cell activity, which is in line with most previous studies assessing the effect of a psychological intervention on this immune parameter conducted in breast cancer patients.25,27,57-59

Thus, although additional research is needed, a relationship between clinical insomnia and host defense mechanisms, particularly on cytokine expression, appears conceivable. However, the clinical relevance of immune changes associated with insomnia treatment in terms of breast cancer progression remains to be demonstrated. Indeed, it is unknown whether immune alterations observed in this study are of a sufficient magnitude to influence breast cancer progression. Even more important, the role of the immune system on cancer progression, particularly in the context of solid tumors such as breast cancer, has yet to be established.2 The modest continuation of immune changes during the follow-up phase of this study also makes us question their potential impact on a long-lasting process such as cancer progression.

The mechanisms through which a psychological treatment of insomnia, such as the one administered in this study, may influence immunity remains uncertain. Overall, there is some indication of a mediating role of improved sleep and reduced psychological distress, but the magnitude of these mediating effects are highly variable across immune parameters. In addition, these variables explain 60% of the absolute change in immune functioning at best. Although not negligible, this suggests that other variables are influential. Other likely mechanisms that should be investigated in future studies include reduced stress level and related endocrine alterations, and increased social support, all of which, unfortunately, were not measured in this study.

The finding showing that decreased fatigue associated with CBT was associated with decreased immune functioning deserves some comment. Although counterintuitive, this finding is consistent with recent literature showing that higher levels of fatigue are associated with immune activation in breast cancer survivors, including increased cytokine activity.60 This is also consistent with this study's findings (see the companion report30) showing no significant superiority of CBT for insomnia compared with the control group for reducing fatigue. Together, these results suggest that factors other than insomnia may influence fatigue in this population and that the relationship between sleep, fatigue, and immune functioning is complex.

To our knowledge, this study is the first to assess the influence of clinical insomnia on immune functioning using an experimental design. Although characterized by several strengths (eg, control of several potential confounders and strict selection of patients), this study is limited by the fact that both groups of patients received the insomnia treatment. A WLC condition is advantageous from an ethical viewpoint but precludes the evaluation of whether immune changes continue, as well as the assessment of their clinical impact on health. Whenever possible, future studies should use a no-treatment control condition and observe patients for a longer period of time to assess cancer- and other health-related (eg, occurrence of infections) outcome variables. In addition, results obtained in this study with breast cancer survivors may not generalize to patients with other types of cancer, although breast cancer shares many characteristics with other cancer types in terms of medical (eg, treatments, prognosis) and psychological aspects.

Because the role of host defense mechanisms on breast cancer progression is debated, future studies conducted in the context of cancer should investigate the impact of clinical insomnia or its treatment on other biologic markers that clearly have been associated with breast cancer progression, such as hormonal factors (eg, estrogen and progesterone levels), angiogenesis, and apoptotic processes.2,61 Indeed, these may constitute potential alternative biologic pathways through which insomnia—or other psychological factors—could influence cancer progression. Alternatively, the impact of treating insomnia on immunologic functioning could also be assessed during the perioperative period; it has been suggested that immunosuppression could increase the risk of metastases, particularly during the perioperative period.62 Nevertheless, the role of clinical insomnia on immune functioning of cancer patients remains extremely relevant from a general viewpoint given that immune functioning defects may lead to other health problems (eg, infectious disease) that may significantly affect the quality of life and even the longevity of cancer patients.2

The authors indicated no potential conflicts of interest.

Table

Table 1. Immunologic Measures (adjusted means and standard errors)

Table 1. Immunologic Measures (adjusted means and standard errors)

VariablesNo.Cognitive-Behavioral Therapy (n = 27)
Waiting-List Control (n = 30)
Pooled Data (n = 57)
Mean95% CIMean95% CIMean95% CI
White blood cells, × 109/L
    Prewaiting306.125.56 to 6.67
    Pretreatment*565.995.42 to 6.576.345.79 to 6.896.175.78 to 6.55
    Post-treatment506.075.48 to 6.666.245.68 to 6.806.155.76 to 6.55
    3-month follow-up436.455.83 to 7.066.656.07 to 7.226.556.14 to 6.95
    6-month follow-up435.935.30 to 6.566.726.13 to 7.316.325.90 to 6.74
    12-month follow-up366.335.64 to 7.036.976.37 to 7.576.656.21 to 7.10
Monocytes, × 109/L
    Prewaiting300.400.04 to 0.46
    Pretreatment*560.380.32 to 0.440.430.37 to 0.480.400.36 to 0.44
    Post-treatment500.400.34 to 0.460.440.38 to 0.500.420.38 to 0.46
    3-month follow-up430.460.39 to 0.530.410.35 to 0.470.440.39 to 0.48
    6-month follow-up430.450.39 to 0.520.470.41 to 0.530.460.42 to 0.51
    12-month follow-up360.460.38 to 0.540.510.44 to 0.570.480.43 to 0.53
Lymphocytes, × 109/L
    Prewaiting301.791.62 to 1.96
    Pretreatment*561.681.49 to 1.872.021.83 to 2.201.851.72 to 1.98
    Post-treatment501.751.56 to 1.952.001.81 to 2.181.881.74 to 2.01
    3-month follow-up431.871.67 to 2.082.202.00 to 2.392.031.90 to 2.17
    6-month follow-up431.961.75 to 2.172.161.96 to 2.362.061.92 to 2.20
    12-month follow-up361.881.65 to 2.122.222.02 to 2.422.051.90 to 2.20
T (CD3+) cells, × 109/L
    Prewaiting281.250.97 to 1.53
    Pretreatment*561.251.04 to 1.461.441.24 to 1.651.351.21 to 1.49
    Post-treatment501.601.37 to 1.821.391.19 to 1.601.491.34 to 1.64
    3-month follow-up431.421.17 to 1.661.451.22 to 1.671.431.27 to 1.59
    6-month follow-up431.471.23 to 1.721.451.23 to 1.681.461.30 to 1.63
    12-month follow-up361.471.19 to 1.761.451.22 to 1.681.461.28 to 1.64
T-helper (CD4+) cells, × 109/L
    Prewaiting280.890.79 to 0.98
    Pretreatment*560.890.78 to 0.991.020.92 to 1.120.960.88 to 1.03
    Post-treatment500.940.83 to 1.051.000.90 to 1.100.970.90 to 1.04
    3-month follow-up431.000.88 to 1.111.070.96 to 1.171.030.96 to 1.11
    6-month follow-up431.020.91 to 1.141.050.94 to 1.151.040.96 to 1.11
    12-month follow-up361.030.90 to 1.161.050.93 to 1.161.040.96 to 1.12
T-cytotoxic/suppressor (CD8+) cells, × 109/L
    Prewaiting280.370.31 to 0.42
    Pretreatment*560.340.28 to 0.400.390.33 to 0.450.370.32 to 0.41
    Post-treatment500.330.27 to 0.400.380.32 to 0.440.360.31 to 0.40
    3-month follow-up430.340.27 to 0.400.410.35 to 0.480.380.33 to 0.42
    6-month follow-up430.400.33 to 0.460.410.35 to 0.470.400.36 to 0.45
    12-month follow-up360.370.30 to 0.450.420.36 to 0.490.400.35 to 0.44
NK (CD16+/CD56+) cells, × 109/L
    Prewaiting270.130.10 to 0.16
    Pretreatment*560.130.10 to 0.160.140.11 to 0.170.130.11 to 0.16
    Post-treatment500.130.09 to 0.170.170.13 to 0.210.150.13 to 0.18
    3-month follow-up430.120.08 to 0.160.160.13 to 0.200.140.12 to 0.17
    6-month follow-up430.110.66 to 0.150.150.11 to 0.190.130.10 to 0.15
    12-month follow-up360.110.07 to 0.160.180.15 to 0.220.150.12 to 0.18
IL-1β production, pg/mL
    Prewaiting291,324.101,028.39 to 1,619.81
    Pretreatment*561,354.08960.42 to 1,747.741,219.11839.22 to 1,598.991,286.591,021.72 to 1,551.47
    Post-treatment501,450.141,034.73 to 1,865.541,630.641,245.39 to 2,015.891,540.391,265.14 to 1,815.63
    3-month follow-up441,830.141,394.92 to 2,265.361,066.87660.32 to 1,473.421,448.501,159.40 to 1,737.61
    6-month follow-up431,669.231,219.62 to 2,118.831,459.551,041.47 to 1,877.631,564.391,265.65 to 1,863.13
    12-month follow-up361,736.991,222.99 to 2,250.981,814.821,390.99 to 2,238.661,775.911,452.57 to 2,099.24
IFN-γ production, pg/mL
    Prewaiting29620.10372.42 to 867.79
    Pretreatment*50684.67414.26 to 955.08588.73362.35 to 815.12636.70469.80 to 803.60
    Post-treatment46812.64539.01 to 1,086.27805.06576.19 to 1,033.93808.85638.40 to 979.30
    3-month follow-up40467.72183.79 to 751.66837.14598.44 to 1,075.84652.43475.07 to 829.79
    6-month follow-up39739.23444.81 to 1,033.65935.42690.36 to 1,180.49837.33654.14 to 1,020.51
    12-month follow-up28869.92544.35 to 1,195.49544.50270.28 to 818.72707.21504.43 to 909.99
NK cell activity (ratio 6.25:1)
    Prewaiting286.505.54 to 7.47
    Pretreatment*536.275.18 to 7.356.325.41 to 7.236.295.59 to 7.00
    Post-treatment476.485.38 to 7.576.936.00 to 7.876.715.99 to 7.41
    3-month follow-up436.745.60 to 7.886.055.06 to 7.046.405.64 to 7.14
    6-month follow-up436.805.62 to 7.996.025.01 to 7.026.415.64 to 7.18
    12-month follow-up366.915.58 to 8.256.295.21 to 7.376.605.75 to 7.45
NK cell activity (ratio 12.5:1)
    Prewaiting287.156.09 to 8.20
    Pretreatment*538.797.57 to 10.027.146.11 to 8.177.977.17 to 8.76
    Post-treatment467.896.62 to 9.167.676.60 to 8.757.786.96 to 8.60
    3-month follow-up438.477.17 to 9.787.296.13 to 8.447.887.01 to 8.74
    6-month follow-up438.947.58 to 10.317.085.91 to 8.258.017.12 to 8.90
    12-month follow-up368.456.88 to 10.027.616.35 to 8.888.037.03 to 9.03
NK cell activity (ratio 25:1)
    Prewaiting288.897.51 to 10.28
    Pretreatment*5311.129.40 to 12.849.087.64 to 10.5110.108.98 to 11.21
    Post-treatment4510.338.56 to 12.109.708.20 to 11.2110.028.87 to 11.17
    3-month follow-up4312.9511.13 to 14.769.237.64 to 10.8211.099.89 to 12.29
    6-month follow-up4313.9912.10 to 15.878.546.93 to 10.1411.2610.04 to 12.48
    12-month follow-up3510.057.82 to 12.278.827.09 to 10.569.448.04 to 10.83
NK cell activity (ratio 50:1)
    Prewaiting2712.0710.22 to 13.92
    Pretreatment*4914.7812.59 to 16.9712.5610.69 to 14.4313.6712.23 to 15.11
    Post-treatment4015.6313.31 to 17.9513.0411.08 to 15.0014.3312.84 to 15.83
    3-month follow-up4216.4814.23 to 18.7412.7110.67 to 14.7414.5913.08 to 16.10
    6-month follow-up4316.1013.76 to 18.4411.969.93 to 13.9814.0312.50 to 15.56
    12-month follow-up3212.299.50 to 15.0911.038.67 to 13.4011.669.85 to 13.48

Abbreviations: NK, natural killer; IL, interleukin; IFN, interferon.

*Can also be called postwaiting for patients in the control condition.

Table

Table 2. Tests of Mediator Effects of Insomnia, Psychological Distress, and Fatigue Indices in the Relationship Between the Insomnia Treatment and Selected Immune Measures

Table 2. Tests of Mediator Effects of Insomnia, Psychological Distress, and Fatigue Indices in the Relationship Between the Insomnia Treatment and Selected Immune Measures

Immune MeasureTotal Time Effects
Direct Time Effects
Indirect (mediator) Effects
↓Insomnia
↓Distress
↓Fatigue
BSE%BSE%BSE%BSE%BSE%
IFN-γ+164.87*50.94100.0+167.6971.5363.0−28.2246.1410.6+47.91*28.4418.0−22.24*20.158.4
IL-1β+309.00*109.33100.0+300.91140.9478.8+44.0282.0611.5−9.4144.782.5−27.6736.957.2
WBC+0.200.16100.0+0.050.2211.4+0.22*0.1447.2+0.060.0913.4−0.13*0.1028.1
Lymphocytes+0.140.07100.0+0.050.1028.9+0.09*0.0655.1+0.010.047.0−0.020.049.0

NOTE. B (and SE) indicates the mean change scores (and SE) estimated with mixed model analyses. Total time effect = direct time effect + indirect effects (mediators). A positive B indicates an increase of this immune variable over time. A negative B indicates a decrease of this immune variable over time. Percentages indicate the percentage of absolute change accounted for by the effect. Effects were computed from the pre- v post-treatment comparison for IFN-γ and IL-1β, and from the pretreatment v 3-month follow-up for WBCs and lymphocytes.

Abbreviations: IFN, interferon; IL, interleukin.

*P <.01.

P <.05.

© 2005 by American Society of Clinical Oncology

Supported in part by an operating grant (MT-14039) and salary support from the Canadian Institutes of Health Research.

Presented in part at the 2nd Annual Conference of the American Psychosocial Oncology Society, Phoenix, AZ, January 2005.

Authors' disclosures of potential conflicts of interest are found at the end of this article.

We thank Catherine Quesnel, Isabelle Giguère, Lucie Casault, Séverine Hervouet, Véronique Dupéré, Aude Caplette-Gingras, Célyne Bastien, Marie-Anne Quesnel, Nancy Roberge, Manon Lamy, Annick Ferland, Chantal Kenney, Julie Dumont, Monique Hamel, Pascal Lebreton, Lucie Vaillancourt, and Paule Rhéaume for their important contributions.

1. Andersen BL, Kiecolt-Glaser JK, Glaser R: A biobehavioral model of cancer stress and disease course. Am Psychol 49::389,1994-404, Crossref, MedlineGoogle Scholar
2. Garssen B, Goodkin K: On the role of immunological factors as mediators between psychosocial factors and cancer progression. Psychiatry Res 85::51,1999-61, Crossref, MedlineGoogle Scholar
3. Kiecolt-Glaser JK, McGuire L, Robles TF, et al: Psychoneuroimmunology: Psychological influences on immune function and health. J Consult Clin Psychol 70::537,2002-547, Crossref, MedlineGoogle Scholar
4. Cohen S, Miller GE, Rabin BS: Psychological stress and antibody response to immunization: A critical review of the human literature. Psychosom Med 63::7,2001-18, Crossref, MedlineGoogle Scholar
5. Cohen S, Rabin BS: Psychologic stress, immunity, and cancer. J Natl Cancer Inst 90::3,1998-4, Crossref, MedlineGoogle Scholar
6. Herbert TB, Cohen S: Depression and immunity: A meta-analytic review. Psychol Bull 113::472,1993-486, Crossref, MedlineGoogle Scholar
7. Savard J, Morin CM: Insomnia in the context of cancer: A review of a neglected problem. J Clin Oncol 19::895,2001-908, LinkGoogle Scholar
8. Morin CM, Ware JC: Sleep and psychopathology. Appl Prev Psychol 5::211,1996-224, CrossrefGoogle Scholar
9. Kiecolt-Glaser JK, McGuire L, Robles TF, et al: Psychoneuroimmunology and psychosomatic medicine: Back to the future. Psychosom Med 64::15,2002-28, Crossref, MedlineGoogle Scholar
10. Born J, Lange T, Hansen K, et al: Effects of sleep and circadian rhythm on human circulating immune cells. J Immunol 158::4454,1997-4464, MedlineGoogle Scholar
11. Dinges DF, Douglas SD, Hamarman S, et al: Sleep deprivation and human immune function. Adv Neuroimmunol 5::97,1995-110, Crossref, MedlineGoogle Scholar
12. Dinges DF, Douglas SD, Zaugg L, et al: Leukocytosis and natural killer cell function parallel neurobehavioral fatigue induced by 64 hours of sleep deprivation. J Clin Invest 93::1930,1994-1939, Crossref, MedlineGoogle Scholar
13. Shearer WT, Reuben JM, Mullington JM, et al: Soluble TNF-alpha receptor 1 and IL-6 plasma levels in humans subjected to the sleep deprivation model of spaceflight. J Allergy Clin Immunol 107::165,2001-170, Crossref, MedlineGoogle Scholar
14. Irwin M, McClintick J, Costlow C, et al: Partial night sleep deprivation reduces natural killer and cellular immune responses in humans. FASEB J 10::643,1996-653, Crossref, MedlineGoogle Scholar
15. Irwin M, Mascovich A, Gillin JC, et al: Partial sleep deprivation reduces natural killer cell activity in humans. Psychosom Med 56::493,1994-498, Crossref, MedlineGoogle Scholar
16. Irwin M, Thompson J, Miller C, et al: Effects of sleep and sleep deprivation on catecholamine and interleukin-2 levels in humans: Clinical implications. J Clin Endocrinol Metab 84::1979,1999-1985, MedlineGoogle Scholar
17. Irwin M: Effects of sleep and sleep loss on immunity and cytokines. Brain Behav Immun 16::503,2002-512, Crossref, MedlineGoogle Scholar
18. Irwin M, Smith TL, Gillin JC: Electroencephalographic sleep and natural killer activity in depressed patients and control subjects. Psychosom Med 54::10,1992-21, Crossref, MedlineGoogle Scholar
19. Cover H, Irwin M: Immunity and depression: Insomnia, retardation, and reduction of natural killer cell activity. J Behav Med 17::217,1994-223, Crossref, MedlineGoogle Scholar
20. Savard J, Miller SM, Mills M, et al: Association between subjective sleep quality and depression on immunocompetence in low-income women at risk for cervical cancer. Psychosom Med 61::496,1999-507, Crossref, MedlineGoogle Scholar
21. Savard J, Laroche L, Simard S, et al: Chronic insomnia and immune functioning. Psychosom Med 65::211,2003-221, Crossref, MedlineGoogle Scholar
22. Irwin M, Clark C, Kennedy B, et al: Nocturnal catecholamines and immune function in insomniacs, depressed patients, and control subjects. Brain Behav Immun 17::365,2003-372, Crossref, MedlineGoogle Scholar
23. Vgontzas AN, Chrousos GP: Sleep, the hypothalamic-pituitary-adrenal axis, and cytokines: Multiple interactions and disturbances in sleep disorders. Endocrinol Metab Clin North Am 31::15,2002-36, Crossref, MedlineGoogle Scholar
24. Kiecolt-Glaser JK, Glaser R: Measurement of immune response, in Cohen S, Kessler RC, Gordon LU (eds): Measuring Stress: A Guide for Health and Social Scientists . New York, NY, Oxford University Press, , pp,1995 213-229 Google Scholar
25. Andersen BL, Farrar WB, Golden-Kreutz DM, et al: Psychological, behavioral, and immune changes after a psychological intervention: A clinical trial. J Clin Oncol 22::3570,2004-3580, Crossref, MedlineGoogle Scholar
26. Gruber BL, Hersh SP, Hall NRS, et al: Immunological responses of breast cancer patients to behavioral interventions. Biofeedback Self-Regul 18::1,1993-22, Crossref, MedlineGoogle Scholar
27. Larson MR, Duberstein PR, Talbot NL, et al: A presurgical psychosocial intervention for breast cancer patients: Psychological distress and the immune response. J Psychosom Res 48::187,2000-194, Crossref, MedlineGoogle Scholar
28. McGregor BA, Antoni MH, Boyers A, et al: Cognitive-behavioral stress management increases benefit finding and immune function among women with early-stage breast cancer. J Psychosom Res 56::1,2004-8, Crossref, MedlineGoogle Scholar
29. Schedlowski M, Tewes U, Schmoll H-J: The effects of psychological intervention on cortisol levels and leukocyte numbers in the peripheral blood of breast cancer patients, in Lewis CE, O'Sullivan C, Barraclough J (eds): The Psychoimmunology of Cancer: Mind and Body in the Fight for Survival? Oxford, United Kingdom, Oxford University Press, , pp 336,1994-348 Google Scholar
30. Savard J, Simard S, Ivers H, et al: Randomized study on the efficacy of cognitive-behavioral therapy for insomnia secondary to breast cancer, Part I: Sleep and psychological effects. J Clin Oncol 23::6083,2005-6096, LinkGoogle Scholar
31. Bastien CH, Vallières A, Morin CM: Validation of the Insomnia Severity Index as an outcome measure for insomnia research. Sleep Med 2::297,2001-307, Crossref, MedlineGoogle Scholar
32. Zigmond AS, Snaith RP: The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand 67::361,1983-370, Crossref, MedlineGoogle Scholar
33. Savard J, Laberge B, Gauthier JG, et al: Evaluating anxiety and depression in HIV-infected patients. J Pers Assess 71::349,1998-367, Crossref, MedlineGoogle Scholar
34. Smets EMA, Garssen B, Bonke B, et al: The multidimensional fatigue inventory (MFI) psychometric qualities of an instrument to assess fatigue. J Psychosom Res 39::315,1995-325, Crossref, MedlineGoogle Scholar
35. Fillion L, Gelinas C, Simard S, et al: Validation evidence for the French Canadian adaptation of the Multidimensional Fatigue Inventory as a measure of cancer-related fatigue. Cancer Nurs 26::143,2003-154, Crossref, MedlineGoogle Scholar
36. Morin CM: Insomnia: Psychological assessment and management . New York, NY, The Guilford Press, 1993 Google Scholar
37. Quesnel C, Savard J, Simard S, et al: Efficacy of cognitive-behavioral therapy for insomnia in women treated for nonmetastatic breast cancer. J Consult Clin Psychol 71::189,2003-200, Crossref, MedlineGoogle Scholar
38. Robinson JP: Handbook of Flow Cytometry Methods . New York, NY, Wiley-Liss, 1993 Google Scholar
39. Chang L, Gusewitch GA, Chritton DBW, et al: Rapid flow cytometric assay for the assessment of natural killer cell activity. J Immunol Methods 166::45,1993-54, Crossref, MedlineGoogle Scholar
40. Kroesen B-J, Mesander G, ter Haar JG, et al: Direct visualisatioin and quantification of cellular cytotoxicity using two colour fluorescence. J Immunol Methods 156::47,1992-54, Crossref, MedlineGoogle Scholar
41. Papadopoulos NG, Dedoussis GVZ, Spanakos G, et al: An improved fluorescence assay for the determination of lymphocyte-mediated cytotoxicity using flow cytometry. J Immunol Methods 177::101,1994-111, Crossref, MedlineGoogle Scholar
42. Kirchner H, Kleinicke C, Digel W: A whole-blood technique for testing production of human interferon by leukocytes. J Immunol Methods 48::213,1982-219, Crossref, MedlineGoogle Scholar
43. Krueger JM, Toth LA: Cytokines as regulators of sleep. Ann N Y Acad Sci 739::299,1994-310, Crossref, MedlineGoogle Scholar
44. Elsasser-Beile U, von Kleist S, Sauther W, et al: Impaired cytokine production in whole blood cell cultures of patients with gynaecological carcinomas in different clinical stages. Br J Cancer 68::32,1993-36, Crossref, MedlineGoogle Scholar
45. De Groote D, Zangerle PF, Gevaert Y, et al: Direct stimulation of cytokines (IL-1 beta, TNF-alpha, IL-2, IFN-gamma and GM-CSF) in whole blood: Comparison with isolated PBMC stimulation. Cytokine 4::239,1992-248, Crossref, MedlineGoogle Scholar
46. Tabachnik BG, Fidell LS: Using Multivariate Statistics (ed 4) . New York, NY, Harper Collins Publishers, 2001 Google Scholar
47. SAS Institute: SAS/STAT User's Guide, Version 8 : Volumes 1, 2, and 3. Cary, NC, SAS Institute, 2001 Google Scholar
48. Keselman HJ, Algina J, Kowalchuk RK: The analysis of repeated measures designs: A review. Br J Math Stat Psych 54::1,2001-20, Crossref, MedlineGoogle Scholar
49. Kirk RE: Experimental design: Procedures for the behavioral sciences (ed 3) . Pacific Grove, CA, Brooks/Cole, 1995 Google Scholar
50. Frigon J-Y, Laurencelle L: Analysis of covariance: A proposed algorithm. Educ Psychol Meas 53::1,1993-18, CrossrefGoogle Scholar
51. Health Canada: Canada's Food Guide to Healthy Eating. Ottawa, Ontario, Canada, Publications Health Canada, 1997 Google Scholar
52. Baron RM, Kenny DA: The moderator-mediator variable distinction in social psychological research: Conceptual, strategic, and statistical considerations. J Pers Soc Psychol 51::1173,1986-1183, Crossref, MedlineGoogle Scholar
53. MacKinnon DP, Lockwood CM, Hoffman JM, et al: A comparison of methods to test mediation and other intervening variable effects. Psychol Methods 7::83,2002-104, Crossref, MedlineGoogle Scholar
54. Benca RM, Quintans J: Sleep and host defenses: A review. Sleep 20::1027,1997-1037, MedlineGoogle Scholar
55. Dickstein JB, Moldofsky H: Sleep, cytokines and immune function. Sleep Med Rev 3::219,1999-228, Crossref, MedlineGoogle Scholar
56. Fawzy FI, Kemeny ME, Fawzy NW, et al: A structured psychiatric intervention for cancer patients. II. Changes over time in immunological measures. Arch Gen Psychiatry 47::729,1990-735, Crossref, MedlineGoogle Scholar
57. Hosaka T, Tokuda Y, Sugiyama Y, et al: Effects of a structured psychiatric intervention on immune function of cancer patients. Tokai J Exp Clin Med 25::183,2000-188, MedlineGoogle Scholar
58. Richardson M-A, Post-White J, Grimm EA, et al: Coping, life attitudes, and immune responses to imagery and group support after breast cancer treatment. Altern Ther Health Med 3::62,1997-70, Google Scholar
59. van der Pompe G, Duivenvoorden HJ, Antoni MH, et al: Effectiveness of a short-term group psychotherapy program on endocrine and immune function in breast cancer patients: An exploratory study. J Psychosom Res 42::453,1997-466, Crossref, MedlineGoogle Scholar
60. Bower JE, Ganz PA, Aziz N, et al: Fatigue and proinflammatory cytokine activity in breast cancer survivors. Psychosom Med 64::604,2002-611, Crossref, MedlineGoogle Scholar
61. Esteva FJ, Hortobagyi GN: Prognostic molecular markers in early breast cancer. Breast Cancer Res 6::109,2004-118, Crossref, MedlineGoogle Scholar
62. Ben-Eliyahu S: The promotion of tumor metastasis by surgery and stress: Immunological basis and implications for psychoneuroimmunology. Brain Behav Immun 17::S27,2003-S36, (suppl 1) Crossref, MedlineGoogle Scholar
Downloaded 1,358 times

COMPANION ARTICLES

No companion articles

ARTICLE CITATION

DOI: 10.1200/JCO.2005.12.513 Journal of Clinical Oncology 23, no. 25 (September 01, 2005) 6097-6106.

Published online September 21, 2016.

PMID: 16135476

ASCO Career Center