Corrigendum: Preventing sleep disruption with bright light therapy during chemotherapy for breast cancer: a phase II randomized controlled trial

Disturbed sleep is one of the most common and distressing complaints among patients with breast cancer, occurring in 30-50% of patients undergoing chemotherapy (Savard & Morin, 2001). Nighttime sleep disruptions, such as difficulty falling asleep, staying asleep, and frequent awakenings, are aggravated in women with breast cancer undergoing chemotherapy (Ancoli-Israel et al., 2006;Berger et al., 2007;Palesh et al., 2010). Patients with cancer also complain of increased daytime napping (Engstrom et al., 1999) described as longer and more frequent daytime naps as treatment progresses (Berger & Farr, 1999;Levin et al., 2005;Wielgus et al., 2009;Young-McCaughan et al., 2003), which has been associated with decreased daytime activity that, in turn, has been found to predict higher cancer-related fatigue (CRF) (Berger & Farr, 1999;Wielgus et al., 2009).Many studies measuring sleep in cancer have used actigraphs, a wrist worn device which measured activity which can be used to estimate sleep and wake. Despite the ability of actigraphy to simultaneously measure both sleep and activity (Berger et al., 2008), relatively few studies have evaluated both outcomes in patients with breast cancer undergoing chemotherapy (Berger et al. 2007;Young-McCaughan et al. 2003).Previous research in our laboratory found that women with breast cancer have decreased daytime light exposure both before and during chemotherapy, with the most pronounced decrease in light exposure during the treatment infusion weeks of chemotherapy (Liu et al., 2005). Synchronized endogenous circadian activity rhythms are related to exposure to diurnal bright light (Kripke et al., 2007); low diurnal illumination levels have been associated with nocturnal sleep dysfunction (Ancoli-Israel et al. 2002;Terman et al. 1995). Sleep and mood disruptions have been successfully treated with morning exposure to increased artificial bright light in other populations, including individuals with winter depression (Rosenthal et all., 1985;Terman and Terman 2005), nonseasonal depression (Al-Karabi and Jubair, 2016), anxiety (Youngstedt and Kripke 2007), and PTSD (Youngstedt et al. 2021). Our laboratory has shown that morning bright light therapy prevents cancer related fatigue from getting worse, prevents circadian activity rhythms from deteriorating and improves quality of life in women undergoing chemotherapy for breast cancer (Neikrug et al. 2012;Ancoli-Israel et al. 2011;Jeste et al., 2013). Morning light therapy has been combined with cognitive behavioral therapy to improve sleep in women undergoing chemotherapy (Bean et al. 2020); however, there are no studies evaluating just bright light therapy on sleep or activity in this group. Thus, we evaluated whether administration of bright light upon awakening in the morning would alleviate the poor nighttime sleep and lower daytime alertness experienced during chemotherapy in women with breast cancer.We conducted a small phase II randomized clinical pilot study comparing bright white light (BWL) therapy to dim red light (DRL) therapy in women diagnosed with breast cancer undergoing four cycles of adjuvant or neo-adjuvant chemotherapy. The study was conducted between July 2005-June 2007Data were collected from the same women reported in previous publications on the effect of light on fatigue, circadian activity rhythms and quality of life (Ancoli-Israel et al., 2011, Neikrug et al., 2012;Jeste et al., 2013). As reported in those studies, 58 women were referred by physicians for the study (see Fig 1). Of those referred, 17 were ineligible after screening and 41 were consented and randomized. Of the 41 randomized, 2 participants (one from each group) dropped out immediately and were not included in the analysis; 8 women from the BWL and 3 women from the DRL dropped during the treatment phase and were included in the analysis. Therefore, data are presented from 39 women (mean age=53.95 yrs., SD=9.06, range=32-70 years).Participants were referred by medical oncologists in the San Diego community or from the UCSD Moores Cancer Center. Inclusion criteria were having a new diagnosis of stage I-III breast cancer and scheduled to receive at least four cycles of adjuvant or neoadjuvant chemotherapy. Exclusion criteria were being pregnant, having metastatic or IIIB (including inflammatory) breast cancer, significant pre-existing anemia, or confounding underlying medical illnesses or any other physiological or psychological impairments that would have limited participation. Breast cancer disease staging was based on the American Joint Committee on Cancer Staging Manual 5th Edition (Greene 2002). Menopausal status was determined using self-report of the occurrence of menses (Rissling et al. 2011).After referral from the oncologist, informed consent, HIPAA, and release of information were obtained by the study coordinator. Pertinent medical information (e.g., stage of disease and estrogen/progesterone receptor status [ER/PR]) was abstracted from each participant's medical record prior to participation in the study.Approval for this study was received from the University of California San Diego Office of IRB Administration and by the UC San Diego Moores Cancer Center's Protocol Review and Monitoring Committee.All women provided written informed consent before participation. The study was registered with the National Institutes of Health ClinicalTrials.gov (Clinical Trials number NCT00478257).Figure 2 shows the study design which included a baseline assessment, treatment randomization prior to the start of chemotherapy followed by daily morning light treatment for four cycles of chemotherapy. After baseline, actigraphy and questionnaires were repeated only during the treatment and recovery weeks of cycles 1 and 4 of chemotherapy. Each chemotherapy cycle was either 2 weeks or 3 weeks as the recommended chemotherapy regimen changed in the middle of our study. Wrist actigraphs were worn for three consecutive 24-hour periods (72-hours) at each of the 5 time-points: prior to the start of chemotherapy (baseline), chemotherapy treatment week of cycle 1 (C1TW), recovery week of cycle 1 (C1RW), chemotherapy treatment week of cycle 4 (C4TW), and recovery week of cycle 4 (C4RW). Actigraphy periods coincided with each participant's scheduled weekday chemotherapy infusions. All questionnaires could be filled out any time during the three days that actigraphy data were collected. The actigraph and the questionnaires were all picked up together.Results of questionnaire data assessing fatigue, mood, quality of life, functional outcome, menopausal status and climacteric symptoms have been previously published (Ancoli-Israel et al., 2011;Jeste et al., 2013;Liu et al., 2009Liu et al., , 2012;;Neikrug et al., 2012;Rissling et al., 2011).Randomization. The randomization sequence was generated by the study statistician using the R statistical software package (http://cran.r-project.org/). A blocked design with a 2:3 allocation to dim red light (DEL; n=16) versus bright white treatment (BWL; n=23) using a block size of 4. Our hypothesis was that BWL would be more beneficial; and therefore, more participants were randomized to BWL to provide a larger sample with this treatment. Both treatments were noninvasive. The study coordinators were blinded to the randomization allocation of participants.Instructions to participants. Each participant was provided with a Litebook®, a demonstration of proper operation, a paper tape measure and digital timer. Participants were instructed how to position the Litebook® and to operate it for 30 continuous minutes immediately upon awakening every day throughout their four cycles of chemotherapy. The goal of the study was described to participants by the study coordinators as an evaluation of two frequencies of light therapy (red or white) for improving sleep and fatigue during chemotherapy.Light. Light was administered via a Litebook 1.2 (Litebook®, Ltd. Alberta, Canada). The Litebook® is a small (6"x5"x1") and lightweight (8 oz.) light box designed to be placed on a table about 18" from the patient's head and within a 45° visual field. As previously published (Neikrug et al., 2012;Ancoli-Israel et al., 2011), light was administered with the Litebook® 1.2 (Litebook, Ltd., Alberta, Canada). The Litebook® utilizes 60 white lightemitting diode (LED) lights with a distribution of energy particularly concentrated in the middle and long wavelengths (Desan et al., 2007) and which mimic the visible spectrum of sunlight (about 1,500 lux) for minimum glare and maximum eye comfort, without emitting ultraviolet (UV) light. Two women randomized into the BWL group reported the light aversive and dropped out during treatment; however, these data are included in the analysis. An identical-appearing device utilizing red LEDs emitting dim red light at <50 lux was used for the comparison DRL group. No participants reported the dim red light aversive.The Litebooks® were modified to include an integrated meter which allowed for monitoring treatment adherence by recording operation time and duration. Partial adherence data were available for 30 participants (BWL n=17; DRL n=13); analysis indicated similar frequency of use (BWL = 55.0%; DRL = 70.2% of days assigned) and duration of use (BWL = 31.5 min [SD=9.89]; DRL = 33.9 min [SD=10.93] per day used) with no significant difference between the groups.Pittsburgh Sleep Quality Index (PSQI). (Buysse et al., 1989) Actigraphy. Wrist actigraphy devices were used for obtaining objective measures of nighttime and daytime sleep as well as for activity levels. Wrist actigraphy measures motion over time by recording the amount of electrical deflection during a fixed interval (e.g., minute by minute) (Ancoli-Israel, et al., 2003;Ancoli-Israel et al., 2015). In the current study, two similar actigraphy devices were used. The Actillume® was used with the first 11 participants (Ambulatory Monitoring, Inc., Ardsley, New York). Actillume® data were analyzed using the Action-3 software program (Ambulatory Monitoring Inc., Ardsley, NY, USA). Actillume® data for 9 participants ( BWL=5; DRL=4) are included in these analyses. The Actiwatch-Light® (Mini-Mitter|Respironics/Philips, Eindhoven, The Netherlands) was used in the remainder of the participants (n=28).Actiwatch-Light® data were analyzed using the Actiware® 5 sleep and activity monitoring software program (Mini-Mitter|Respironics). Activity sensitivity threshold was set to medium. Both devices record continuous acceleration data on the non-dominant wrist using a battery-operated microprocessor that senses motion with a piezoelectric beam and detects movement in all three axes. As previously published, device equivalency was evaluated by comparing data collected by paired devices worn simultaneously for 72-hours by eight healthy adult volunteers (Ancoli-Israel, et al., 2014: Liu et al., 2013a;Liu et al., 2013b). The software-scored sleep/wake data based on the two types of activity count were highly correlated (both r's>0.85, both p's<0.0001), therefore, these variables were deemed equivalent for the purpose of this study. Data from 39 participants (BWL=23, DRL=16) are included in analyses.Actigraphic sleep variables were derived from a mean of three continuous sleep and wake (night/day) periods using 1-minute epochs. Self-report via sleep log was used to edit actigraphy data and determine daytime and nighttime sleep and wake periods. Nighttime variables included: nighttime average activity counts per minute, sleep percentage (%sleep), nighttime total sleep time (TST) and nighttime total waketime (TWT). Daytime variables included: daytime average activity counts per minute, mean nap duration (mNAP) number of daytime naps (nNAP) and total nap time (TNT). A daytime sleep episode, or nap, was defined as any period of 10 or more minutes of consecutive actigraphic inactivity (i.e., sleep) during the period between final out of bedtime in the morning and into bedtime the following night.Descriptive statistics were calculated for the entire sample as well as separately for the two treatment groups. Group differences were assessed with t-tests at baseline for possible confounders (i.e., demographic variables, clinical characteristics, and chemotherapy regimen). Variables that significantly differed between the treatment groups at a 0.05 significance level were controlled for in the inferential analysis.Linear mixed-effects models were used for analyzing changes of subjective sleep quality, objective activity count and sleep/wake variables before and during chemotherapy, with group, time and group-by-time interaction included as fixed covariate effects. Baseline was the reference time point and the DRL group was the reference group. Each of the outcome variables were modeled separately. If a significant group, time or group-x-time interaction was found, further post-hoc tests were conducted using appropriate contrasts: between group differences at each time point, and/or within group changes from Baseline to the other time points. Linear mixed-effects models and restricted maximum likelihood methods (Diggle, Liang, & Zeger, 1996) were employed for analyzing and comparing sleep and activity variables for each treatment group. This paradigm relies on the "missing at random" assumption (Diggle, 2002) and allows for modeling partial data where the number of measures per person could vary and participants with missing time points could still be included in the analysis. Thus, mixed model protects from a "completers only" bias.Table 1 shows the sociodemographic characteristics of our sample. There were no significant differences between the treatment groups in age, BMI, race, income, education, marital status, ER/PR status, or stage of disease.At baseline, the BWL group had significantly less TST than the DRL group (p=0.01), thus the baseline TST was adjusted in all linear mixed-effects models. No other group differences were found at baseline (both p's>0.2 for %sleep and TWT).While controlling for baseline differences, a significant group-by-time interaction was found for TST at both C4TW (p=0.042) and C4RW (p=0.012) (Fig. 3A). Compared with baseline, the BWL group had significant increases in TST at C4TW and C4RW (p's<0.03), whereas the DRL group had no significant changes in TST at these time points. Significant group-by-time interactions were found for %sleep and TWT at C4RW. Compared with baseline, the DRL had a significant decrease in %sleep (p<0.05) and a significant increase in TWT (p<0.05) at C4RW, whereas the BWL had no significant changes in these variables at these time points. Compared with baseline, in the DRL group, TNT increased significantly at C4TW (p=0.0003) and but changes during C4RW were not significant. In the BWL group there were no significant changes in TNT.Significant group-by-time interactions for mNAP were found at C1TW (p<0.03), C4TW (p<0.05) and C4RW (p<0.05) (Fig 4C). Compared with baseline, mNAP increases significantly at C1TW (p<0.0003) and C4TW (p<0.05) and non-significantly at C4RW (p<0.05) in the DRL group, whereas compared with baseline, mNAP had a small (non-sgnificant) increase at C1TW, and small (non-significant) decreases at C4TW and C4RW in the BWL group. As shown in Fig. 5A, activity counts during the nighttime sleep period did not differ between groups at baseline (p=0.16). At C4RW a significant group-by-time interaction was found (p=0.047). Compared with baseline, at C4RW the DRL group had a significant increase in nighttime sleep period activity (p=0.033), whereas the BWL group showed a non-significant decrease in nighttime sleep activity count. No other groupby-time effect was found for activity during the night period.As shown in Fig. 5B, activity counts during the daytime wake period did not differ between groups at baseline (p=0.35). A significant group-by-time interaction was observed at C4TW (p=0.013). Compared with baseline, at C4TW the DRL group had significantly less activity counts (p<0.001), whereas the BWL did not have a significantly change in activity count. A similar pattern was observed during C1TW; compared with baseline, activity counts decreased in the DRL group (p but did not change significantly in the BWL group. the group-by-time interaction was not significant at C1TW, for other There were no significant differences in the or between the BWL group and the DRL group at baseline and also no significant group-by-time interactions for the or during either cycle the BWL group, compared to baseline, there were significantly lower in three (i.e., in subjective sleep quality, sleep sleep during C4RW, however, the daytime dysfunction increased (i.e., daytime during both weeks of cycle 4 (both the compared to baseline, the subjective sleep quality decreased during the recovery weeks of both cycles (i.e., sleep quality but the use of increased during the treatment weeks of both cycles of this study that morning bright white light administered daily during chemotherapy to women with breast cancer of nighttime sleep and sleep quality and daytime the weeks of chemotherapy the weeks of the women in both treatment groups more and longer the recovery week of cycle both groups to the levels. by the the of chemotherapy in less sleep at night and more and longer naps during the day in the women in the DRL group women in the BWL group showed an increase in nighttime sleep and a to levels of were observed in sleep no significant group by time interaction was observed by the of cycle 4 chemotherapy, compared to baseline, the BWL group reported in nighttime sleep quality sleep quality, sleep duration and sleep of daytime dysfunction This of daytime during the cycle of treatment be not only to but also to the of cancer treatment. The that the DRL group reported sleep quality during the treatment weeks of both cycles be by the increase in sleep use et al. the objective sleep and subjective sleep quality that the bright white light in less of to the on significant changes in the amount of activity both during the sleep period and the wake period were et al. showed that there is between night and day as measured by actigraphy during chemotherapy, which both sleep at night and wake during the day ( actigraphic activity counts during the wake period and low counts during the sleep period has been associated with higher et al., et al., quality of life et al., and lower levels of depression et al., et al., in patients with a more circadian pattern of activity and sleep has also been of less cognitive in women with breast cancer (Ancoli-Israel, et al. 2021). In the current study, women to dim red light had decreased activity during chemotherapy treatment weeks of cycle 1 and cycle as be during chemotherapy, those to bright white light had no significant changes in daytime activity compared to the sleep those in the BWL group showed less activity than those in the DRL group. data that bright white light also the women from the in activity experienced during chemotherapy. activity levels during the wake period are low et al., increases in activity could sleep et al., and circadian (Youngstedt et al., for this study was the prior that women undergoing chemotherapy receive less bright light exposure as treatment particularly in the days following chemotherapy and that this decrease is associated with fatigue and sleep (Liu et al. 2005). reported data from this sample that bright light therapy fatigue (Ancoli-Israel et al. and of both the circadian activity (Neikrug et al. and subjective quality of life during chemotherapy et al. 2013). The current not only a of of sleep and activity in the bright white light group but also in daytime sleep and nighttime activity compared to that the most sleep reported in the study are the of circadian by light (Neikrug et al., light the circadian by in the which with the via a as well as from the to the and and 2010). of bright light on alertness and et al., also have to nighttime as well as napping in the study. whether the observed of bright light therapy was to the effect of to the in circadian activity rhythms or other be determined from this there have been a few other studies the effect of bright light treatment on sleep in cancer et al. et al. et al. et al. et al. to our is the first randomized controlled the effect of bright light therapy on sleep both and and activity during chemotherapy in women with breast et al. found a effect on sleep in a similar using a modified behavioral therapy that included both nighttime and daytime sleep however, the in the treatment group was limited to an in subjective sleep quality and measured at night sleep was not these are that both nighttime and daytime sleep using a behavioral treatment be during chemotherapy, the of objective in sleep also that be the significant group-by-time for of the variables, the clinical significance of of these was bright light has been shown to be highly in fatigue, circadian rhythms and quality of life in these same women (Neikrug et al. 2012;Ancoli-Israel et al. 2011;Jeste et al., the sleep low and total wake time at all time bright light with other such as cognitive behavioral treatment for (Bean et al., or et al., have with clinical of the current study include the randomized controlled and in the of a baseline prior to chemotherapy in to data during include the of both subjective and objective measures of sleep and the of the mixed model statistical analysis which allowed for (i.e., missing data at the of "completers only" there are also to the study. The first is the small sample a larger sample such as the found in the DRL group have been significant. this was a study to provide II data for a larger randomized the activity be with as we did not Our was on sleep for which wrist actigraphy is a measure (Ancoli-Israel, et al., 2003;Ancoli-Israel et al., 2015). of activity and is in A be the period of actigraphy data This period was to both and to time for baseline data as the time period between and the start of chemotherapy was less than one the breast cancer chemotherapy group dim red light showed and during chemotherapy, particularly in daytime and inactivity during the day during cycle 4. The bright white light group however, showed significantly less and were less and more during the day at cycle 4 a ability to from the of chemotherapy. studies are to these however, the study that morning bright an behavioral at least of nighttime sleep and daytime activity and alertness in women undergoing chemotherapy for breast

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.