A Review Of The Circadian Rest-Activity Rhythm, A Potential Safety Pharmacology Article Review Example

A Review Of The Circadian Rest-Activity Rhythm, A Potential Safety Pharmacology Article Review Example

The “circadian timing system” has been linked to the wellness and improvement of survival rate in cancer patients. Recent research shows how the circadian timing system is affected by chemotherapy in mice. The study tries to answer the main question: what is the relationship between the disruption of circadian rhythm and adverse events in individual cancer patients? The study also aims to quantify the treatment-induced CTS changes and to characterize the rhythm recovery dynamics of cancer patients.
The study samples include 49 patients who are candidate for saving chemotherapy regimens and medical treatment of their cancer disease. Specifically, the criteria include those who were diagnosed to have locally-advanced or metastatic solid cancer, at least 18 years old, had a physician rated performance status less than 3 and the expected life expectancy is at least 6 months. Other criteria include; actigraphy records taken at least 2 weeks after the end of the previous therapy and recovery from many acute clinical or hematological toxicities.  A Review Of The Circadian Rest-Activity Rhythm, A Potential Safety Pharmacology Article Review Example.The study design involves repeated measurements and longitudinal designs to determine CTS patterns of cancer patients who received a specific amount of body clock related chemotherapy dose. Patients were subjected to wrist actigraphy four times for in 3-4 consecutive days each before, during, immediately and after chemotherapy. Some of the parameters taken include the estimated activity within and out of bed, the “autocorrelation coefficient,” amplitude, within and among daily variability.
Results revealed that a specific chronotherapy protocol may disrupt circadian in in-out bed activity (I<O 97.5%). The result is also significant when all the rest-activity rhythm factors are disrupted (p<0.05). The average values that were estimated were also close to baseline values. There is also a significant result suggesting that there is a higher risk of “fatigue” at p=0.028 and “body weight loss” at p=0.05 during the disruption the circadian pattern of chemotherapy. Of the total number of patients, 9.5% exhibit no change when chronotherapy was introduced to the patient. Only 14.3% showed improvement while 31% of the total patients showed changes and full recovery. Patients who exhibited a sustained deterioration accounted to 45% of the total population. In conclusion, minimizing disruption of the body clock through a personalized dosage of chronotherapy is related with improved clinical tolerability of cancer patients. A Review Of The Circadian Rest-Activity Rhythm, A Potential Safety Pharmacology Article Review Example.


Psychosocial symptoms often cluster together, are refractory to treatment, and impair health‐related quality of life (HR‐QoL) in cancer patients. The contribution of circadian rhythm alterations to systemic symptoms has been overlooked in cancer, despite a causal link shown under jet lag and shift work conditions. We investigated whether the circadian rest‐activity rhythm provides a reliable and objective estimate of the most frequent patient‐reported outcome measures (PROMs).


Two datasets were used, each involving concomitant 3‐day time series of wrist actigraphy and HR‐QoL questionnaires: EORTC QLQ‐C30 was completed once by 237 patients with metastatic colorectal cancer; MD Anderson Symptom Inventory (MDASI) was completed daily by 31 patients with advanced cancer on continuous actigraphy monitoring, providing 1015 paired data points. Circadian function was assessed using the clinically validated dichotomy index I < O. Nonparametric tests compared PROMs and I < O. Effect sizes were computed. Sensitivity subgroup and temporal dynamics analyses were also performed.


I < O values were significantly lower with increasing symptom severity and worsening HR‐QoL domains. Fatigue and anorexia were worse in patients with circadian disruption. The differences were both statistically and clinically significant (P < 0.001; d ≥ 0.33). Physical and social functioning, and global quality/enjoyment of life were significantly better in patients with robust circadian rhythm (P < 0.001; d ≥ 0.26). Sensitivity analyses validated these findings.


Objectively determined circadian disruption was consistently and robustly associated with clinically meaningfully severe fatigue, anorexia, and interference with physical and social functioning.A Review Of The Circadian Rest-Activity Rhythm, A Potential Safety Pharmacology Article Review Example.  This supports an important role of the circadian system in the determination of cancer patients’ HR‐QoL and symptoms that deserves therapeutic exploitation.

Keywords: actigraphy, Circadian, patient‐reported outcome, quality of life, symptom
Go to:

Currently, one in eight adults carries wearable “well‐being” activity monitors,1 with some 50 million such units been sold in the United States in 2016.2 This positive perception of e‐Health devices in the general population represents an opportunity for implementing objective measures of physiology and behavior complementing the assessments of symptoms and health‐related quality of life (HR‐QoL), especially in cancer patients.3 Indeed, systemic psychosocial symptoms are usually subjectively rated using validated questionnaires evaluating patient‐reported outcome measures (PROMs).4More specifically, fatigue, sleep disturbance, depression, and anorexia represent the most frequent psychosocial complaints of cancer patients.5 Extensive research has shown that cancer patients tend to display multiple systemic symptoms that often cluster together.6Fatigue, anorexia, and affective disorders can also arise as a consequence of anticancer treatment, suggesting shared physiopathological mechanisms.7 Such systemic ailments also reveal the disruption of those body clocks that time behavioral and cellular activities along the 24‐hour cycle, for example, as a consequence of jet lag or shift work.891011All living beings, from unicellular organisms to humans, are endowed with endogenous biological clocks that enable living organisms to anticipate cyclic environmental changes and coordinate physiological events.12 The mammalian circadian timing system (CTS) is hierarchically organized and temporally controlled, and it coordinates several physiological processes, at whole‐body, cellular, down to molecular, levels.1314 In particular, sleep‐wake cycles, physical and mental performance, as well as appetite, are modulated along the 24 hours by the CTS.1516 As altered patterns have been described for several circadian rhythms in cancer patients,151718 we hypothesized that systemic symptoms would be more severe in patients with circadian disruption. A Review Of The Circadian Rest-Activity Rhythm, A Potential Safety Pharmacology Article Review Example. Two independent datasets were used to test this hypothesis. Based on the convenience of its noninvasive methodology and previous reports in smaller cohorts showing an association with fatigue and sleep problems,19202122232425 we selected the circadian rest‐activity rhythm as measured by wrist actigraphy.2627 Wrist actigraphy has been validated as an objective biomarker of circadian function.28 Finally, objective actigraphy data were correlated with selected subjective PROMs, including systemic symptoms and HR‐QoL domains.

Go to:

2.1. Study populations

For this study, we used datasets from two patient cohorts, involving different PROMs tools and methodologies. The original data were available for both objectively assessed circadian rest‐activity rhythm (wrist actigraphy)28 and subjectively rated symptoms as well as HR‐QoL (with validated questionnaires) from the same patients. For both patient cohorts, approval had been obtained from the appropriate ethical review boards, and patients had provided signed informed consent.293031

2.1.1. Cohort #1

The first set was composed of patients with metastatic colorectal cancer, who were not at time of monitoring having anticancer treatment and had a WHO performance status of 0, 1, or 2. They had participated to either a monocentric study from May 1994 to January 1997 (Chronotherapy Unit, Department of Medical Oncology, Paul Brousse University Hospital, Villejuif, France)31 or to a companion study of an international randomized trial involving nine institutions in four countries, from August 1999 to February 2002.29

Patients in both studies underwent wrist actigraphy monitoring for 72 consecutive hours (Mini‐motionlogger, Ambulatory Monitoring Inc., Ardsley, NY, USA) and completed the European Organization for Research and treatment of Cancer Quality of Life Core (EORTC QLQ‐C30 v2.0) questionnaire.32 This internationally validated 30‐item questionnaire incorporates eight symptoms, five functioning domains and one global QoL scale. All scores were transformed to a 0‐100 scale, according to the recommended EORTC procedures. A Review Of The Circadian Rest-Activity Rhythm, A Potential Safety Pharmacology Article Review Example. For the symptom scales, low scores corresponded to mild symptoms, whereas for the global QoL and its domains scales, low scores indicated poor functioning.32 In the analysis, we selected the systemic symptoms (fatigue, anorexia, sleep problems, pain), global QoL, and the functioning domains (physical, role, and social), corresponding to the items in the M.D. Anderson Symptom Inventory (MDASI).33

2.1.2. Cohort #2

The second set included patients with histologically proven advanced or metastatic cancer requiring medical treatment. The patients had participated to the pilot project on multidimensional tele‐monitoring from home performed at the Chronotherapy Unit, Department of Medical Oncology, Paul Brousse University Hospital, Villejuif, France, between April 2012 and July 2013, within the framework of the inCASA European project (FP7).30 The patients were equipped with a home‐based platform for multidimensional tele‐monitoring over at least 30 consecutive days. This remote surveillance included continuous wrist actigraphy (Micro‐motionlogger, Ambulatory Monitoring Inc., Ardsley, NY, USA) and once‐daily completion of an electronic version of the MDASI questionnaire33 using an interactive screen. This 19‐item validated questionnaire evaluates the severity (from 0 to 10) of 13 core symptoms and of their interference with six activities of daily living. Large scores for each item indicate severe symptoms.33 Patients were monitored while being treated with chemotherapy, as indicated according to their medical condition.30 Hence, this data‐dense study provided dynamic patterns of circadian rest‐activity rhythm and PROMs. As both circadian rest‐activity rhythm and symptom severity on chemotherapy are not stationary and present temporary variations,34 we did not pool all data from a single patient over the monitoring span (exceeding 30 days). Instead, we analyzed actigraphy data over 72 consecutive hours, with 3‐day sliding windows and a 1‐day shift, throughout the time series in each patient, and used individual daily data in MDASI items scores, as previously described.30 Hence, this dataset provided a larger amount of data than the actual number of patients. For the analysis, we selected the systemic symptoms (fatigue, anorexia, sleep disturbance, pain) and the interference items (general activity, work, relations with others, and enjoyment of life), corresponding to the items from the EORTC questionnaire. A Review Of The Circadian Rest-Activity Rhythm, A Potential Safety Pharmacology Article Review Example.

Among both cohorts, there were no uncontrolled metabolic, endocrine, or autoimmune diseases and no symptomatic brain metastases (the details are provided in the original papers293031).

Figure S1 displays the study flowchart.

2.2. Wrist actigraphy

The wrist‐worn accelerometers used in both cohorts were manufactured by the same company (Ambulatory Monitoring Inc., Ardsley, NY, USA), which provided also the dedicated analytical software (Action 4). In the second cohort study, the patients downloaded and tele‐transmitted the rest‐activity data collected over the past 24 hours using the home Internet platform.30 For both cohorts, the epoch length for data collection was set at 1 minute, according to common practice.28 The actigraph collects and stores the number of wrist accelerations per minute, across the three axes, from the nondominant arm. The pattern of accelerations over time is then analyzed to compute pertinent parameters for assessing circadian rest‐activity rhythm, over 72 consecutive hours, as recommended.28

2.3. Statistical analyses

We selected the dichotomy index I < O as the most clinically relevant actigraphy parameter, based on prior studies from others and ourselves.29303135363738 I < Ois the percentage of activity counts per minute when the patient is in bed at night with values lower than the median activity count when the patient is out of bed during the day.39 Hence, it can range from 0% to 100%. In case of restful sleep at night and lively activity during the day, a robust and prominent circadian rhythm is present, and I < Owill be close to 100%.39 To categorize patients with circadian disruption or not, we used the cut‐off point for I < O of 97.5%, as previously identified and validated.3637 Thus, when I < O was lower or equal to 97.5% we estimated that circadian rest‐activity rhythm disruption was present, whereas this rhythm was deemed maintained when I < O was greater than 97.5%.

Summary statistics were computed to describe the distribution of I < O values (median and interquartile range) and of PROMs (means, SD, and SEM). First, we categorized the PROMs items into terciles, and compared the distributions of I < O among the terciles with the Jonkheere‐Terpstra test. Secondly, we defined two categories of patients using the previously established cut‐off point of 97.5% for I < O as a marker of circadian disruption.A Review Of The Circadian Rest-Activity Rhythm, A Potential Safety Pharmacology Article Review Example.  The distribution of the EORTC or the MDASI items were compared between the two groups, using an independent sample t test. We also evaluated the effect size of the difference in PROMs scores between the two groups by computing Cohen’s d, with a threshold for clinically meaningful difference set at d ≥ 0.25. Additionally, we assessed the clinical relevance of the absolute differences based on previously identified thresholds: 10 points for the EORTC questionnaire and one point for the MDASI scale, respectively.4041 We used also the nonparametric Mann‐Whitney U test to compare questionnaires items according to I < O category, as sensitivity analysis. For cohort #1, we performed subgroup analysis according to sex, PS, and age, using the same methodology. For cohort #2, we performed additional comparisons of the dynamic patterns of PROMs and I < O. Thus, we computed the differences in I < O between each day and the previous one, with a sliding window approach. The distribution of changes in selected PROMs was compared in each of the three subgroups defined by the terciles of the changes in I < O (improved, stable, worsened) with Wilcoxon signed‐rank test.

Moreover, Spearman’s rank correlations between I < O and selected PROMs indices were computed for each cohort. Finally, we built a multivariate linear regression model with global quality of life (for cohort #1) or interference with enjoyment of life (for cohort #2) as dependent variables, and all the other selected PROMs of each questionnaire and I < O as independent variables, to assess the objective, additional information about HR‐QoL provided by I < O. Analyses were performed using PASW v24 (SPSS, IBM Inc., Chicago, IL, USA) and Stata v14 (StataCorp LLC, College Station, TX, USA) software packages. The threshold for statistical significance was set at P ≤ 0.005, correcting for multiple comparisons. A Review Of The Circadian Rest-Activity Rhythm, A Potential Safety Pharmacology Article Review Example.

Go to:

3.1. Study populations

Study cohort #1 included 237 patients with metastatic colorectal cancer, who completed the EORTC QLQ‐C30 questionnaire and underwent 3‐day wrist actigraphy recording (Table S1). Study cohort #2 included 31 patients, mostly suffering from advanced or metastatic gastro‐intestinal malignancy and having 1015 valid dyads of 3‐day wrist actigraphy recordings and daily completion of the MDASI questionnaire, at the intermediate day of the 3‐day actigraphy sliding window (Table S1). Altogether, nearly 90% of the patients in either population had a performance status of 0 or 1, despite advanced disease (Table S1).293031

3.2. Descriptive statistics

Table S2 provides mean and SD values for EORTC symptom scales and quality of life domains (range, 0 to 100) in cohort #1, and for MDASI items (range, 0 to 10) in cohort #2 (Table S2).

The cut‐off points for the terciles of PROMs are detailed in Table S3.

In both populations, the distribution of I < O values was skewed toward high values, as in previous reports3842 (Figure S2). Median values, in both cases, were close to 97.5%, formerly reported as a clinically meaningful cut‐off point3637: 96.9% [1st and 3rd quartiles: 93.6%‐99.1%] for cohort #1, and 98.0% [95.8‐99.0] for cohort #2 (Figure S1). Thus, the proportion of instances with circadian disruption (ie, with I < O ≤ 97.5%) was 54.9% in cohort #1 and 44.4% in cohort #2. A Review Of The Circadian Rest-Activity Rhythm, A Potential Safety Pharmacology Article Review Example.


3.3. Comparative analyses of wrist actigraphy monitoring and questionnaires

In cohort #1, I < O significantly decreased with increasing severity of fatigue (P < 0.0001), anorexia (P < 0.0001), pain (P < 0.0001), and sleep trouble (P = 0.003) (Figure 1A). In contrast, I < O significantly increased with greater values of global quality of life (P < 0.0001), physical (P < 0.0001), and social (P < 0.0001) functioning, but not role (P = 0.02) functioning (Figure 1B). In cohort #2, significantly lower I < Ovalues were observed with gradually more severe fatigue and anorexia, as well as interference with enjoyment of life, activity, relations with others, and work (all P < 0.0001), whereas differences were not significant for sleep disturbance (P = 0.56) and pain (P = 0.009; Figure 1C,D).

An external file that holds a picture, illustration, etc.
Object name is CAM4-7-4396-g001.jpg
Figure 1

Median (and interquartile range) I < O values in the subgroups defined by the terciles of the PROMs items from the EORTC (panels A and B) and MDASI (panels C and D) questionnaires: blue, first; gray, second; red, third. In all cases, the higher the tercile, the more severe the symptom, except for the EORTC QLQ‐C30 domains (panel B), in which the higher the tercile, the better the quality of life and the functioning. In all cases, stars indicate < 0.0001. Other nonsignificant P values are detailed in the Section 3

The comparison of PROMs as a function of circadian disruption (I < O ≤ 97.5%) or robustness (I < O > 97.5%) yielded similar results. A Review Of The Circadian Rest-Activity Rhythm, A Potential Safety Pharmacology Article Review Example. Thus, in cohort #1, patients with circadian disruption complained of statistically more severe fatigue (P < 0.0001), anorexia (P < 0.0001), and pain (P < 0.0001), yet only a nonsignificant trend (P > 0.005) was found for sleep trouble (P = 0.009). In cohort #2, fatigue (P < 0.0001) and anorexia (P < 0.0001) were also rated as significantly more severe when I < O ≤ 97.5%, whereas sleep disturbance (P = 0.61) and pain (P = 0.02) were not.

In cohort #1, global quality of life (P < 0.0001), physical functioning (P < 0.0001), and social functioning (P < 0.0001) were rated as significantly poorer by patients with I < O ≤ 97.5% as compared to those with higher I < O values, whereas role functioning (P = 0.04) was not. In cohort #2, instances with circadian disruption were significantly associated with greater interference with enjoyment of life, activity, relations with others, and work (all P < 0.0001). The associated effect sizes were of intermediate magnitude in both populations for the statistically different items (Table 1). Figure 2displays the mean (±SEM) values for both populations for selected corresponding symptoms (panel A) or functioning/interference items (panel B). A Review Of The Circadian Rest-Activity Rhythm, A Potential Safety Pharmacology Article Review Example.