GW280430A: Pharmacodynamics and Potential Adverse Effects

The three reports by our group in the April (2004) issue of Anesthesiology1–3on the new ultrashort-acting nondepolarizing relaxant GW280430A (AV430A) have evoked invited commentary by Caldwell4and observations by Geldner and Blobner. Both anticipation and caution were expressed by these commentators regarding the possible future clinical performance of AV430A.

To add to the discussion, I would offer a rather optimistic viewpoint based on additional data in preparation for publication. I am certainly biased because of my closeness to the development of AV430A. Nevertheless, I do believe that future results will strengthen the candidacy of AV430A as a replacement for succinylcholine or as an excellent alternative.

AV430A is a representative of the new class of nondepolarizing relaxants, which we have called asymmetrical mixed-onium chlorofumarates.5These compounds are inactivated by two entirely chemical (nonenzymatic) mechanisms both in vivo  and in vitro : cysteine adduction and pH-sensitive hydrolysis.1,5,6The cysteine adduction reaction is notably manipulatable.5,6In AV430A, both reactions are fast (combined T^1/2β approximately 1–2 min) such that the duration of block is ultrashort, similar in duration to or even shorter than succinylcholine in cats, dogs, and three different primate species.1,3 

Notable characteristics of this series include high potency, easy antagonism by anticholinesterases, and rapid antagonism by cysteine (see last paragraph). AV430A is particularly nonvagolytic,1an important contrast vis-à-vis  rapacuronium, and does not block sympathetic ganglia.1AV430A is approximately four times weaker than mivacurium with respect to release of histamine in the rhesus monkey.1These autonomic data suggest that the only possible side effect of AV430A that might appear in the clinical dose range is a rather mild tendency to cause minor symptoms and signs suggestive of slight degrees of histamine release, exemplified by facial erythema, brief decrease in blood pressure, and increase in heart rate.

AV430A and its related series were first studied in 1997. Despite encouraging results in human volunteers,2the development process has been delayed. First, a corporate decision by Glaxo Wellcome (Research Triangle Park, NC) to discontinue the development of products for the field of anesthesia, and then the merger of Glaxo Wellcome and Smith Kline, halted the project for 4 yr until the AV430A series was acquired by Avera Pharmaceuticals (San Diego, CA) in 2002. In 2003, a second study of a new formulation of AV430A in volunteers was completed and reported in 2004. The kinetic and safety data are encouraging.7,8 

The failure of rapacuronium and its removal from the market, as a result of the side effect of bronchospasm, will place greater scrutiny on the possible airway effects of all new relaxants. An ultrashort-acting agent such as AV430A, which will no doubt be given for tracheal intubation, will bear the greatest scrutiny. Bronchospasm after rapacuronium, although noted in clinical trials, was far more serious in clinical practice with rapacuronium than foreseen before its release. This is the crux of the problem: Minimal effects of AV430A on the airway must be shown to provide convincing data regarding the safety of AV430A. Caldwell4and Geldner and Blobner both allude to this.

Following is a discussion of some of the data regarding AV430A that are already published, which suggest that AV430A should be safe as far as airway effects are concerned. Minimal airway effects have already been found in dogs in doses up to 50 × ED⁹⁵.3The nonvagolytic properties of AV430A found in cats1mean that it is unlikely to have any blocking effects on M2 or M3 receptors in the human airway. This indicates minimal possibility of bronchospasm on this basis, in contrast with the severe responses noted after rapacuronium. Jooste et al.  10have shown that mivacurium has minimal interactions with M2 and M3 receptors9where the actions of rapacuronium are clearly demonstrable. Because mivacurium, which is a predecessor of AV430A, has approximately the same large safety ratio (> 25) as does AV430A regarding lack of interaction with muscarinic receptors in cats,1the imbalanced blocking property on muscarinic receptors in the human airway, which is the likely mechanism of rapacuronium-induced bronchospasm,10is a most unlikely clinical scenario in the case of AV430A.

What about histamine release? Both Caldwell4and Geldner and Blobner express concern about this, but in the rhesus monkey, the safety ratio of AV430A for this side effect is approximately four times greater than that of mivacurium.1This results in the following comparison in humans: mivacurium, at 2.5–3 × ED⁹⁵(0.20–0.25 mg/kg), causes greater symptomatology of histamine release when injected over 15 s11than does AV430A when injected over only 5 s at 3–4 × ED⁹⁵(0.5 mg/kg).2Since the original study in human volunteers by Belmont et al.  2in 1998, AV430A has been reformulated. This reformulation has further improved the safety ratio for histamine release to approximately 4 × ED⁹⁵in humans.7As a reminder, this “safety ratio” is defined as the dose required to cause an average decrease in blood pressure of 10%, divided by the ED⁹⁵for neuromuscular blockade: ED Hist/ED⁹⁵.1 

Both Caldwell4and Geldner and Blobner caution that anesthesiologists might increase dosage to cause faster onset and thereby increase the possibility of histamine release by AV430A. We have already compared the onset and duration of AV430A in the human larynx and the thumb vis-a-vis  succinylcholine, rapacuronium, rocuronium, mivacurium, and cisatracurium. The data are not yet published but suggest that the onset of AV430A is as fast as that of succinylcholine and faster than those of the others. So why administer AV430A at a dosage higher than 0.4–0.5 mg/kg? Onset of block after AV430A does not get any faster at dosages higher than this and is probably circulation limited, as pointed out by Caldwell.4The onset is fast in all species—dogs, cats, monkeys, and humans.1–3Consequently, Geldner and Blobner, who believe that “GW280430A may be an ultrashort-acting rather than a rapid-onset muscle relaxant” are most likely incorrect. AV430A is clearly ultrashort and very rapid in onset in all studies.1–3,figure 1shows a mechanomyograph recording of the response to AV430A (0.4 mg/kg) in the adductor pollicis in a healthy human volunteer during nitrous oxide–oxygen–fentanyl–propofol anesthesia. The pattern of block seen in figure 1was noted in every volunteer subject of the more than 100 humans treated so far. The dose of 0.4 mg/kg is approximately 3 × ED⁹⁵.7 

Doses up to 5 × ED⁹⁵caused minimal side effects.7A 5-s bolus dose of AV430A (0.4 mg/kg) was given at the arrow. Two control train-of-four (TOF) responses are followed by the elicited twitch in the thumb at 0.15 Hz. Twitch is abolished within approximately 80 s. Recovery begins at approximately +6 min (time scale at top). The time scale changes at and after this point. At +8 min, TOF shows appearance of T⁴, and T¹is at approximately 25% of control. At +10 min, TOF is 45% and T¹is 75% of control. At +12 min, TOF is 85% and T¹is 95% of control. At +13.5 min, TOF is 95% of control (control TOF is at the far left). The time from 5% T¹to a TOF of 90% is 6 min. Heart rate (upper record) and blood pressure (lower record) show no change.

Time and experience will tell, as clinical studies in patients are undertaken, whether early data from volunteers accurately predict the performance of AV430A in practice. The dosage recommended to achieve certain clinical endpoints, e.g. , intubation of the trachea within 60 s, among others, must be defined by these future studies.

In volunteer studies thus far, at dosages as high as 0.8–0.9 mg/kg (approximately 6 × ED⁹⁵or 2 × intubating dosage) where AV430A is given as a rapid (5 s) bolus, the manifestations of histamine release after AV430A are rather mild (facial flushing and brief decrease in blood pressure, not requiring treatment).7There has been no bronchospasm. This suggests that clinicians could give these very large dosages safely, particularly by injecting them a little more slowly (such as over 15 s). The side effect of histamine release as caused by AV430A, because it is four times weaker than it is in mivacurium,1may constitute a minor concern in future clinical practice. Time again will tell. Only after a couple of years of experience in thousands of administrations will the pattern be fully defined.

Nevertheless, we can be reassured by this data. If, as Kopman et al.  12have suggested, a dose of 2.0 or 2.5 × ED⁹⁵of AV430A is enough for good to excellent intubating conditions within 60 s (this dose would be in the range 0.3 ± 0.05 mg/kg), there may very well be minimal side effects.

The chemical pathways of inactivation of AV430A are, in my opinion, its most promising feature. The chemical breakdown will ensure no prolonged neuromuscular blockade. Problems with atypical pseudocholinesterase will not be at issue. Cysteine, given intravenously as a “reversal drug” or “rescue agent,” will rapidly inactivate AV430A. Complete recovery from 100% twitch inhibition can be induced in monkeys with exogenous cysteine within 1–2 min.13In the future, anesthesiologists may have the choice of spontaneous recovery (12–14 min) from AV430A via  endogenous cysteine or induced recovery by giving additional cysteine, e.g. , in case of an airway emergency. Cysteine, given within 2–3 min after injection of AV430A, should abolish complete paralysis within 1–2 min.13The latter treatment with cysteine may shorten the total duration of action in humans to an estimated 5 min to return of full neuromuscular function such as cough, normal vital capacity, and head lift.

Weill Medical College of Cornell University/New York Presbyterian Hospital, New York, New York. jjsavare@med.cornell.edu