The Association of Plasma and Cerebral Spinal Fluid Cytokines with Daytime Sleepiness Among Children with Acute Lymphoblastic Leukemia

Introduction:

Fatigue and daytime sleepiness (DS) are understudied among children with acute lymphoblastic leukemia (ALL). Past studies have identified pro-inflammatory cytokines, such as TNFα and interferon-inducible protein measured in serum as potential mediators of fatigue and sleep disturbances in other populations. Cytokines measured in cerebral spinal fluid (CSF) may correlate more closely with DS than those measured in serum. Yet, CSF measurements of cytokines are rare because they are difficult to obtain. ALL patients undergo lumbar puncture as part of routine exams, thus presenting an opportunity to measure CSF levels of cytokines without additional procedures. The aim of our study was to examine the associations of plasma and CSF cytokines with patient-reported fatigue among children with ALL. Our hypothesis was that higher levels of cytokines and soluble cytokine receptors would be associated with greater patient-reported DS.

Patients and Methods:

This study included 16 children aged 4-11 years with ALL who had cytokine and DS measures available. The cytokine IL-6, and cytokine receptors IL-1 R1 & R2, TNF R1 & R2 were measured in plasma and/or cerebral spinal fluid at up to three time points during maintenance chemotherapy regimens, approximately 3 months apart. DS was assessed at each of these same time points using the Pediatric Daytime Sleepiness Scale (PDSS), which is completed by the child and has possible values between 0 and 32 points. Higher scores on this scale indicate greater DS, with a score of 17 or greater indicating clinically relevant DS. We used repeated measures linear regression models to examine the association of each cytokine with PDSS scores, adjusting for age at each visit.

Results:

The mean age at study entry was 6.1 years (SD=2.4), and 47% of patients were male. The mean PDSS score was 20.4 points (SD=6.1, range 4 to 29), and 79% of the patients had a PDSS score ≥17. Table 1 shows the age-adjusted associations of each cytokine with PDSS score. Each 100 pg/mL increase in CSF TNF R2 was associated with a 2.13 (SE=0.76) point higher score on the PDSS (p=0.01). In general, higher cytokine levels in both plasma and CSF were associated with higher PDSS scores, although no other associations reached statistical significance.

Conclusions:

We found that the majority of ALL patients undergoing maintenance chemotherapy experienced clinically relevant DS. Higher CSF levels of TNF R2 were associated with greater patient-reported DS, which supports results from past studies that measured TNFα in plasma. This association is biologically plausible because TNF R2 mediates the many metabolic effects of TNF that can include fatigue and sleep. We are in the process of evaluating other cytokines and cytokine-receptors on sleep. Our findings suggest that the complex relation of fatigue, daytime-sleepiness, and CSF cytokines should be evaluated in a larger patient population, and that targeting chronic stimulation of the TNFα pathway may help to mitigate daytime sleepiness in ALL patients.

^a Estimates are from repeated linear regression models, adjusted for age at each visit.