RheothRx (Poloxamer 188) Injection for the Acute Painful Episode of Sickle Cell Disease: A Pilot Study

NO EFFECTIVE treatment will terminate the acute painful episodes of sickle cell disease (SCD). Current treatment is primarily supportive and includes measures such as vigorous hydration, heat packs, oxygen, antibiotics, analgesics, and blood transfusions.1-5 Investigational agents such as urea sodium and forms of cyanate and cetiedil citrate initially showed promise for treating the acute painful event but were subsequently found to be too toxic or were only marginally effective.6-10 Prednisone was studied during acute painful events in children and appeared to shorten the duration of the event; however, the effectiveness was limited by a rebound of painful events after discontinuation of the drug.11 Bone marrow (BM) transplantation provides a curative approach; however, it is limited by lack of suitable BM donors and a 9% mortality rate.12-14 More recently, innovative therapies that increase the production of fetal hemoglobin such as hydroxyurea, butyrate analogs, and erythropoietin have been studied. Hydroxyurea is the sole agent that has been shown to decrease the severity of the disease by decreasing hemolysis and reducing the incidence of acute painful events.15-23 Although these agents represent a significant improvement in the overall therapy of SCD, they do not represent a treatment for patients with an ongoing acute painful event.

It is generally accepted that painful episodes and the end organ damage associated with SCD are caused by microvascular occlusion and tissue ischemia resulting from complex interactions between the sickle erythrocytes, endothelium, platelets, plasma clotting factors, and certain mediators of inflammation.24-30 Accordingly, an agent that alters these interactions and restores blood flow would be beneficial. RheothRx (Glaxo Wellcome Inc, Research Triangle Park, NC) Injection is a nonionic block copolymer surfactant composed of blocks of hydrophobic polyoxypropylene and hydrophilic polyoxyethylene that improves microvascular blood flow by lowering viscosity and adhesive frictional forces.31-36 The mechanism of action for these effects is not well understood; however, it is hypothesized that the hydrophobic core of the molecule binds hydrophobic regions of cells, thereby leaving the hydrophilic chains free to interact with the surrounding media. Consequently, RheothRx provides a hydrated, poorly compressible barrier that appears to block hydrophobic adhesive interactions (cell-cell, cell-protein, and protein-protein) in the blood. As a result, there is a reduction in blood viscosity, erythrocyte aggregation, adhesion to vascular endothelium, and an improvement in microvascular blood flow.30-40 

In phase I studies, a 72-hour infusion of RheothRx was generally well tolerated in seven pain-free SCD patients at doses up to 30 mg/kg/h (CytRx Corp, Atlanta, GA, unpublished data).

These preliminary clinical and preclinical studies led us to conduct a phase II pilot study of RheothRx to evaluate its safety and efficacy in adult SCD patients experiencing an acute painful episode.

Study patients.Fifty patients were enrolled between July 13, 1992 and March 31, 1994. SCD patients who were relatively healthy, presented with an acute painful episode within 4 to 18 hours of the onset, and required treatment with parenteral analgesics were eligible for study entry. Inclusion and exclusion criteria are shown in Table 1. The protocol was reviewed and approved by the institutional review boards at each study center and written informed consent was obtained from each patient before study entry.

Study design.This was a randomized, double-blind, placebo-controlled trial conducted at six clinical centers. The protocol was standardized across all centers and routinely monitored for compliance. Patients were randomized to poloxamer 188 (formulated as RheothRx Injection) or placebo in a 1 to 1 ratio (Table 2). Clinical trial materials (RheothRx, placebo), centrally located at Glaxo Wellcome, Inc, were numerically prelabeled in ascending order, packaged as patient-specific boxes, then distributed for storage in pharmacies at each study site for use at enrollment. Study medication was administered as a continuous, two-stage, intravenous infusion over 48 hours (60-minute loading dose of 300 mg/kg followed by a 47-hour maintenance infusion of 30 mg/kg/h). The minimum required duration of hospitalization was 3 days (2 days for the infusion and 1 day for postinfusion observation and assessments).

Study medication.Study medication was packaged by the sponsor in 500-mL glass bottles. Each bottle of RheothRx Injection (a clear, colorless solution) contained 75 g of poloxamer 188 (150 mg/mL) in a buffered saline solution. Placebo consisted of the vehicle for RheothRx Injection. The glass bottles were covered with aluminum foil to maintain blinding of all parties throughout the treatment period.

Concurrent therapy.Parenteral analgesics allowed during the study were meperidine, morphine, and hydromorphone. Oral analgesics were restricted to nonsteroidal anti-inflammatory drugs, meperidine, acetaminophen, aspirin, codeine, oxycodone, and/or any combination of these agents. The appropriate analgesic therapy was administered at the discretion of the investigator according to patient need.

Safety and efficacy.Patient safety was assessed by monitoring adverse experiences, standard laboratory tests, physical examinations, vital signs, and electrocardiograms. Duration of painful episode was the primary efficacy endpoint in this study (Table 3); pain intensity, total analgesic use, and days of hospitalization were secondary efficacy endpoints. Although not a study objective, the incidence of recurrent painful events was also determined.

Statistical analysis.Because the treatment groups differed with respect to intensity of baseline pain, each of the efficacy variables (duration of painful episodes, duration of hospitalization, average pain intensity, and analgesic use) was analyzed using both an unadjusted analysis and an analysis that adjusted for baseline pain intensity. Body weight was also included as a covariate in analyzing analgesic use.

Duration of painful episodes (hours) and duration of hospitalization (days) were analyzed using a log-rank (generalized Savage) test for the unadjusted analysis and a proportional hazards model for the adjusted analysis. Pain intensity was evaluated by a 5-point scale (0 = no pain, 0.5 = residual soreness, 1 = mild pain, 2 = moderate pain, 3 = severe pain). Average pain intensity over 168 hours (ie, the 2-day dosing period plus the 5-day posttreatment period) was derived by calculating the area under the curve formed by plotting the pain intensity score at each assessment from the baseline assessment to 168 hours after initiation of the study drug, and then dividing this area under the curve by 168. If patients did not have pain intensity scores through 168 hours, the score for the last available pain assessment was carried forward to 168 hours. Linear models were used to compare the treatment groups. A similar derivation and analysis was done for the average pain intensity over 72 hours.

To compare the treatment groups with respect to quantity of analgesics given, the doses of all analgesics given (including those not allowed by the protocol) were converted into morphine equivalent units (MEU).5 The total MEU/kg given to the patient from initiation of study drug through the end of the acute painful episode or hospital discharge, whichever came first, were analyzed. A Wilcoxon rank sum test was used for the unadjusted treatment comparison because the assumptions for a linear model were not justifiable.

A rank analysis of covariance41 was done to adjust the analgesic use analysis for baseline pain intensity and body weight. In this analysis MEU and body weight were first ranked across treatment groups. The ranked MEU values were then regressed on all covariables other than treatment (ie, regressed on the ranks for body weight and an indicator variable for baseline pain intensity). Treatment groups were then compared with respect to the residuals from this regression by a Mantel-Haenszel mean score test.

All statistical tests were two-sided, and P values less than .05 were considered statistically significant. The analysis of length of crisis was the primary efficacy endpoint, and all other efficacy endpoints (length of hospitalization, analgesic use, and pain intensity) were secondary endpoints. No adjustment was made for multiple comparisons as regards the secondary endpoints because this was a pilot study and the analyses done are for descriptive purposes only.

Study population.Of the 50 patients enrolled in the trial, 28 were randomized to receive RheothRx and 22 to receive placebo. The proportion of patients contributed by each center are listed in Table 2. The two treatment groups were comparable with respect to the distribution of age, sex, weight, race, and hour from onset of the painful episode to start of study drug (Table 2). There was an imbalance in the baseline pain score between the treatment groups; severe pain was present in 39% of patients who received RheothRx and 64% of patients who received placebo. This apparent imbalance must be interpreted with caution. Before study entry all but four patients (all in the RheothRx group) received parenteral analgesics, which may have altered their baseline pain scores. This imbalance has been addressed in the adjusted analyses of the efficacy endpoints.

In the efficacy analyses, three subgroups of patients were considered. Subgroup 1, with 49 patients, was an “intent-to-treat” subgroup that included all patients for whom efficacy data were available. Subgroup 2, with 45 patients, excluded patients with a study drug infusion duration of less than 24 hours. Subgroup 3, consisting of 31 patients, excluded patients who did not receive the full dose of study drug (ie, 48-hour infusion) or for whom the end-of-painful episode time was estimated (in most cases by adding 12 hours to the time of administration of the last parenteral analgesic dose). Sixteen patients from subgroup 3 were excluded from efficacy analysis because of protocol violations (ie, drug administration errors or incomplete pain assessments), two patients withdrew consent, and one withdrew because of pain at the injection site after 15 minutes of infusion. Safety data were analyzed in all 50 patients.

Efficacy.Data were analyzed for all three subgroups described above for each of the four main efficacy endpoints, except pain intensity, which was analyzed only in subgroup 3 because only this subgroup had pain intensity scores monitored until resolution of the painful episode. Results are shown in Tables 3 and 4. The results of the unadjusted analyses are discussed below. Covariate adjustment for baseline pain intensity had little effect on the P values; however, covariate adjustment for baseline pain and body weight somewhat increased the P values for the analysis of analgesic use. Median duration of painful episodes (Table 3) was reduced in the RheothRx group by 13 hours in subgroup 1 (P = .147), by 28 hours in subgroup 2 (P = .097), and by 36 hours in subgroup 3 (P = .020). Duration of hospitalization (Table 3) was reduced 1 to 2 days in the RheothRx arm of all three subgroups, but these results were not statistically significant. There was a threefold to five-fold reduction in use of analgesics (Table 3) in the RheothRx arm. The reductions were statistically significant in subgroup 2 (P = .037 and P = .048) and subgroup 3 (P = .014 and P = .014) for all analgesics and parenteral analgesics used, respectively. As shown in Table 4, average pain intensity from baseline to 72 hours was significantly lower (P = .034) in the RheothRx group (1.1) compared with that for the placebo group (2.0). The average pain intensity over the interval of baseline to 168 hours was also significantly lower (P = .049) in the RheothRx group (1.4) compared with the placebo group (0.8). Ten of 24 (42%) of the RheothRx-treated patients and 8 of 18 (44%) of placebo-treated patients had a recurrent painful episode by the 1- to 3-week postinfusion follow-up assessment.

Safety.RheothRx was well tolerated. Adverse experiences were similar in both groups, with the most frequent events being headache, nausea, injection site pain, abdominal pain, vomiting, and constipation (Table 5). Most of these events were either mild or moderate in intensity. In addition, one adverse experience considered serious and attributable to study medication was reported; a 20-year-old man with mild underlying renal dysfunction (baseline creatinine 1.5 mg/dL) had a transient increase in serum creatinine concentration during infusion of RheothRx (peak concentration = 2.7 mg/dL). No treatment was required, and the creatinine returned to baseline by the time of the follow-up assessment.

Our results show that RheothRx administered as a 48-hour continuous infusion may be of benefit to SCD patients experiencing acute painful episodes. Administering of RheothRx was associated with a statistically significant reduction in the use of analgesics and pain intensity scores. Whether the imbalance in baseline pain scores between the RheothRx and control group affected the outcome of this study is difficult to interpret. Covariate adjustment for baseline pain had little effect on P values; however, a slight increase in P values was observed when adjusting for baseline pain and body weight for the analysis of analgesic use. However, narcotics were dosed and titrated based upon patient self-reported pain intensity scores and not by body weight. RheothRx administration was also associated with a reduction in the painful episode duration and days of hospitalization; however, these did not achieve statistical significance. Although not statistically significant, examination of the trend in results from the three subsets of patients suggest a dose-related effect of RheothRx. Comparison of the length of painful episodes in the intent-to-treat group, the group excluding patients who had infusion for less than 24 hours, and the group excluding patients who had infusion for less than 48 hours shows the duration of painful episodes was 67 hours, 60 hours, and 44 hours, respectively. Interestingly, the duration of crisis remained essentially unchanged (80 hours, 88 hours, and 80 hours, respectively) in the corresponding subsets of the placebo group. These results suggest efficacy is related to either the total dose or duration of RheothRx administration.

RheothRx was well tolerated with no significant divergence between treatment and placebo groups with respect to adverse experiences. However, the one patient with mild underlying renal dysfunction who had a transient increase in serum creatinine may be important in view of recent results from clinical studies investigating RheothRx in acute myocardial infarction. In two large multinational studies, a dose-related increase in serum creatinine was observed following high doses of RheothRx, especially in elderly patients with preexisting renal disease.42 43 The relevance of these observations to the use of RheothRx in acute painful episodes of SCD is unknown.

Currently, there are no therapies for the acute painful episode other than supportive measures and the use of narcotics analgesics. An agent that can be administered during ongoing painful events and that results in decreased pain and analgesic use may offer the benefit of reducing the complications of narcotic analgesics such as constipation, nausea, itching, sedation, drug tolerance, dependence, and the abstinence syndrome. More importantly, if RheothRx shortens the duration of the painful episode, it may contribute to the preservation of end organ function and reduce the overall morbidity associated with SCD. Shortening painful events and decreasing hospital stay should also result in important cost savings. Finally, if RheothRx is effective, it may also favorably alter the course of other clinical situations associated with vasoocclusion of SCD such as priaprism, acute chest syndrome, and stroke.

Based on this study, it appears appropriate to conduct a larger trial to test the efficacy of this agent for managing of the acute painful episode of SCD. Our understanding of the pathophysiological events that lead to painful episodes and of the possible mechanisms of action of RheothRx suggest that this agent should probably be used in full dose (subgroup 3) as early as possible in the evolution of the painful event rather than delayed until the pain is well-established, as was done in this trial.

The authors thank the study coordinators Linda Norfleet and Ruth Gibson (Memphis, TN); Arleen Anderson (Jackson, TN); Barbara Tomlinson and Rebecca Murgatroyd (Gainesville, FL); Louise Dorn and Nassin Talischy (Chicago, IL); and Johnny Stanfield (Augusta GA).