Effect of Chronic Opioid Therapy on Pain and Survival in a Mouse Model of Sickle Cell Disease

Pain is one of the major comorbidities of sickle cell disease (SCD) requiring chronic opioid therapy (COT). However, COT has been associated with reduced survival and opioid-induced hyperalgesia (OIH), which can exacerbate the chronic pain and increased mortality already inherent to SCD. It is challenging to examine pain, opioid use and survival in populations with SCD due to significant heterogeneity in multiple clinicopathologic factors. To determine the effect of COT we used transgenic HbSS BERK sickle mice, which show pain characteristics similar to those observed in patients with SCD, including sensitivity to thermal and mechanical stimuli, increased opioid requirement, and higher levels of circulating substance P and tryptase (Tran et al., Blood 2017). Additionally, similar to female patients with SCD, female BERK sickle mice show more pain than males. Therefore, we performed a randomized double-blind placebo-controlled trial to examine the effect of COT on pain and survival in female homozygous HbSS BERK (sickle) and HbAA BERK (control) mice, expressing >99% human sickle hemoglobin and normal human hemoglobin A, respectively. Mice were injected subcutaneously each day with either morphine sulfate at a starting dose of 20 mg/kg, which was increased to 25 mg/kg, 30 mg/kg, 35 mg/kg, and 40 mg/kg after weeks 12, 18, 28, 30, and 38 respectively or placebo (equal volume of saline) until the end of survival. Dose of morphine was selected based on the effective analgesic dose for these mice (Kohli et al., Blood 2010). Mechanical-, thermal- (heat and cold), and musculoskeletal/deep hyperalgesia were analyzed in all mice biweekly before and after drug treatments. We used Kaplan-Meier and log-rank test with Sidak correction for survival; 2-way repeated measures analysis of variance with Bonferroni's correction to compare hyperalgesia between time points among control and sickle mice and Cox proportional-hazards regression analysis for the association of hyperalgesia with survival.

We observed significantly decreased survival of saline-treated sickle mice compared to saline-treated control mice (P = 0.0009). Compared to saline, morphine treatment led to a significant decrease in survival in control mice (P = 0.035) but not in sickle mice (P > 0.05). Furthermore, we did not observe an association between hyperalgesia and survival in either control or sickle mice. However, we discerned a significant increase in mechanical, cold, and heat hyperalgesia in control mice after 4 weeks of morphine treatment (P < 0.02, compared to day '0') which continued to increase up to 12 weeks (P <0.05 compared to week 4). Similarly, in sickle mice we observed an increase in mechanical hyperalgesia after 4 weeks (P <0.02, compared to day '0') of morphine treatment which continued to increase up to 12 weeks (P < 0.0001). These data suggest COT leads to OIH in both control and sickle mice. It is noteworthy that morphine treatment continued to show an analgesic effect over the course of 12 weeks in spite of an increase in hyperalgesia in both groups of mice. Thus, mice did not develop tolerance to morphine analgesia. Since mice were treated under uniform conditions, the effect of chronic morphine treatment could be observed without any confounding factors. Our findings in control mice recapitulate the clinical observations that COT is associated with reduced lifespan in non-sickle patients. COT lead to OIH in sickle mice but provided analgesia without causing tolerance or reducing survival. OIH may contribute may exacerbate a vicious cycle of chronic pain and opioid use in SCD. Differences exist between humans and mice including morphine metabolism between the two but due to several similarities of HbSS BERK sickle mice with clinical pain characteristics and biology, these observations have a significant translational potential following clinical trials for optimizing the treatment of pain in SCD. These data highlight the critical need to develop analgesic strategies devoid of OIH.