HONOLULU, HI—When co-administered with pegfilgrastim, naproxen simultaneously reduces cancer-treatment-related bone pain and sleep disturbance, according to results of a secondary analysis reported during the American Pain Society’s 31st Annual Scientific Meeting.

“This treatment combination has the potential to minimize the damaging effects of pegfilgrastim on quality of life and promote chemotherapy adherence,” noted Jennifer S. Gewandter, PhD, University of Rochester Cancer Center Community Clinical Oncology Program, Rochester, NY, and colleagues.

The Phase 3 trial randomly assigned 510 patients at 17 sites to receive naproxen 500mg two times daily or placebo on the day of pegfilgrastim administration. Mean age was 55.6 years; 86% were female. The majority (68%) of the patients had breast cancer and 10% had lung cancer.

Using a visual analog scale, patients rated their sleep disturbance on a scale of 0–10 prior to treatment with pegfilgrastim and rated both sleep disturbance and pain severity, also on a scale of 0–10, for 5 days following treatment with pegfilgrastim.

The naproxen group had a 22% decrease in sleep disturbance, the investigators found. Mean posttreatment sleep disturbance score was 3.74 [95% CI: 3.33, 4.15] in the placebo group and 2.92 [2.51, 3.32] in the naproxen group. No difference in sleep disturbance prior to treatment was observed. “NSAIDs have proven analgesic efficacy but no efficacy in sleep disturbance prevention outside of painful contexts,” they noted.

Using structural equation modeling to assess the mediation effect of pain changes on naproxen-induced changes in sleep disturbance, they found that naproxen intervention was significantly predictive of decreased pain (path coefficient (PC) = -0.143 [-0.240, -0.045]); in addition, changes in pain significantly predicted changes in sleep disturbance (PC = 0.180 [0.123, 0.236]). The mediation effect of pain on sleep disturbance was significant (indirect effect -0.026 [-0.048, -0.009]); however, the direct effect of naproxen on sleep was not (PC = -0.067 [-0.125, 0.008]).

“Results were similar when sleep disturbance was modeled as a mediator of pain; therefore, no definitive cause and effect relationship can be established statistically,” they reported.