Abstract

Background: Since 2012, a growing list of case reports has emerged describing the occurrence of a TTP-like state following intravenous (i.v.) abuse of extended release oxymorphone (Opana® ER). The pathophysiology resulting from intravenous exposure to Opana® ER has demonstrated a spectrum of severity with unclear etiology, and approaches to treatment/management have ranged from early plasma exchange to supportive care alone. As a result these cases have spurred a collaborative investigation to better understand mechanisms of toxicity. Given the temporal relation between the reformulation of the drug product to achieve extended release and the onset of reported cases, we investigated the inert ingredient mixture as a possible causal factor.

Methods: Guinea pigs were used as an animal model to understand the hematopathologic and nephrotoxic potential of the inert ingredient mixture of Opana® ER, which was provided by Endo Pharmaceuticals under a research collaborative agreement. Animal studies were approved by Institutional Animal Care and Use Committees and conducted in adherence with NIH guidelines. Animals were studied over various i.v. dosing schemes (single injections of 0.1, 0.3, or repeated dosing of 0.3mg/kg (5 doses at 1.5hr intervals)) with either the entire inert ingredient mixture ("PEO+") or high molecular weight polyethylene oxide alone ("PEO")-the major constituent of the inert ingredient mixture. The dose range was based on estimates of the inert ingredient mass introduced during the adulteration and injection of the drug product. Additional in vitro and cell-based assessments of coagulation (thromboelastography, thrombin generation) and ADAMTS13 function were carried out to better understand in vivo observations.

Results: Intravenous administration of the solubilized inert ingredient mixture in guinea pigs resulted in dose-dependent pathologic changes in blood parameters and kidney function. Hematological changes included a dose dependent increase in free plasma hemoglobin (Hb), decreased hematocrit and platelet count. Acute kidney injury was characterized by dose-dependent renal cortical iron deposition, plasma creatinine increase, distal convoluted tubular injury (NGAL mRNA induction), and glomerular damage (hemoglobinuria) (Table 1). At high doses, red blood cell aggregation and renal hypoxia (HIF-1 accumulation) was observed. These observations were similar following the administration of PEO alone, and occurred independent of changes to ADAMTS13 and its cleavage of Von Willebrand Factor. PEO, when spiked into blood in vitro, does not directly modulate coagulability as indicated by global assays of coagulation.

Conclusions: The present animal study suggests that i.v. exposure to the inert ingredients included in Opana® ER can trigger hematologic responses resembling elements of TTP, namely intravascular hemolysis and acute renal injury/failure. However, the renal injury observed in the guinea pig model is distinct from classic TTP, which involves spontaneous platelet clumping and subsequent microvascular thrombosis; instead, the nephrotoxicity likely arises from a multifactorial process encompassing renal tubular injury and glomerular damage, arising secondary to hemolysis, extracellular Hb/heme/iron mediated oxidative tissue damage and compromised tissue oxygenation.

Table 1.
Guinea pig dose group
(mg of PEO+/kg) 
Cmax of plasma free Hb [mg/ml] AUC0-24h of plasma free Hb [mg/ml x h] hematocrit at 24h [%] platelet count at 24h [K/µl] NGAL induction** (renal cortex) plasma creatinine at 24h [µmol/L] hemoglobinuria at end of study [mg/ml] iron deposition (renal cortex)
[µg/g tissue] 
Control 0.8 ±0.1 (n=4) 16.2 ±4.8 (n=4) 36 ±1.7 (n=5) 205 ±11.6 (n=5) 1 ±0.3 (n=3) 87.0 ±9.3 (n=4) 0.1 (n=1) 33 ±3.5 (n=3) 
0.1 single dose 1.4 ±0.6 (n=11) 20.5 ±5.75 (n=11) 35 ±3.1 (n=5) 157 ±20.5 (n=5) 1 ±0.5 (n=3) 104 ±41.3 (n=4) 0.6 (n=2) 49 ±7.3 (n=3) 
0.3 single dose 2 ±0.3 (n=13) 36.5 ±5.1 (n=13) 31 ±1.1 (n=6) 135 ±11.2 (n=6) 1.6 ±0.4 (n=3) 145 ±36.7 (n=4) 1.2 (n=2) 153 ±2.5 (n=3) 
0.3 multi dose (5 injections at 1.5h intervals)* 5.1 ±1.3 (n=5) 87.8 ±15.8 (n=4) 28 ±0.8 (n=4) 62 ±12.4 (n=4) 24 ±0.8 (n=4) 421.1 ±33.8 (n=4) 9.8 ±5 (n=3) 139 ±9.5 (n=4) 
Guinea pig dose group
(mg of PEO+/kg) 
Cmax of plasma free Hb [mg/ml] AUC0-24h of plasma free Hb [mg/ml x h] hematocrit at 24h [%] platelet count at 24h [K/µl] NGAL induction** (renal cortex) plasma creatinine at 24h [µmol/L] hemoglobinuria at end of study [mg/ml] iron deposition (renal cortex)
[µg/g tissue] 
Control 0.8 ±0.1 (n=4) 16.2 ±4.8 (n=4) 36 ±1.7 (n=5) 205 ±11.6 (n=5) 1 ±0.3 (n=3) 87.0 ±9.3 (n=4) 0.1 (n=1) 33 ±3.5 (n=3) 
0.1 single dose 1.4 ±0.6 (n=11) 20.5 ±5.75 (n=11) 35 ±3.1 (n=5) 157 ±20.5 (n=5) 1 ±0.5 (n=3) 104 ±41.3 (n=4) 0.6 (n=2) 49 ±7.3 (n=3) 
0.3 single dose 2 ±0.3 (n=13) 36.5 ±5.1 (n=13) 31 ±1.1 (n=6) 135 ±11.2 (n=6) 1.6 ±0.4 (n=3) 145 ±36.7 (n=4) 1.2 (n=2) 153 ±2.5 (n=3) 
0.3 multi dose (5 injections at 1.5h intervals)* 5.1 ±1.3 (n=5) 87.8 ±15.8 (n=4) 28 ±0.8 (n=4) 62 ±12.4 (n=4) 24 ±0.8 (n=4) 421.1 ±33.8 (n=4) 9.8 ±5 (n=3) 139 ±9.5 (n=4) 

data represented as mean ±SEM, n = number of animals

*endpoint 24h, all other groups 48h

**Fold induction over baseline

Disclosures

Wong:Haemonetics Corporation: Research Funding.

Author notes

*

Asterisk with author names denotes non-ASH members.