Amicus fail dents Pompe pipeline

Amicus investors celebrated last year when Sanofi’s follow-on Pompe project failed to show superiority over Lumizyme, the French group's incumbent enzyme-replacement product, in the pivotal Comet trial. A more cautious reading of the setback would have been apt.

The pivotal Propel study of AT-GAA, Amicus’s rival Pompe contender, has missed its primary endpoint, it was revealed yesterday; like Sanofi, Amicus pointed to several other measures of disease progression that it says prove its project to be more effective than Lumizyme. What regulators make of all this is the critical issue , and with a decision due on Sanofi’s avalglucosidase alfa in May, this will soon become apparent.

Amicus remains convinced that AT-GAA is approvable, and intends to complete a rolling US BLA in the second quarter. The 34% drop in the company’s stock this morning shows that investors are less optimistic. Even if the product does make it to market its commercial prospects have likely been seriously dented, and the sellside swiftly moved to slash sales forecasts today.

The disappointment is yet another flashing red warning against reading too much into data generated in early, open-label studies. A phase I/II study of AT-GAA, in 32 patients, found that treatment-naive patients added 63 metres to a six-minute walk test (6MWD), with those switching from Lumizyme adding 42 metres over baseline readings. As a result, AT-GAA was heralded as something of a revolution coming for Pompe patients.

The equivalent measures in Propel came nowhere close to this result. Switch patients improved by almost 17 metres, a result that was at least noticeably better than those in the Lumizyme control cohort, who logged no improvement. Those naive to the ERT added only 10.9 metres, however, and in a further damaging blow this was virtually identical to the 11.1 metres in the control arm.

Propel used 6MWD in the whole population for its primary endpoint, and here the difference between the two cohorts (20.8 vs 7.2 metres) failed to demonstrate that AT-GAA was statistically superior to Lumizyme. This renders all further analyses exploratory only, and despite Amicus claiming “clinically significant” improvements on other measures the trial must be considered a bust.

One of these measures, the lung function test forced vital capacity (FVC), is considered very important in Pompe patients, who mostly die of respiratory failure. 

Effectiveness on this endpoint won Lumizyme approval, and indeed Sanofi used FVC as the primary endpoint in Comet. That trial was considered disappointing because avalglucosidase alfa, also called NeoGAA, only managed to demonstrate non-inferiority to Lumizyme, failing on the statistical superiority measure.

AT-GAA appeared to arrest patients’ respiratory decline, although a breakdown of results by naive and experienced patients shows very little benefit for the former group. Executives speaking on an investor call last night variously blamed this finding on an unexpected Lumizyme result, patient heterogeneity and small numbers.

But the last factor here – the sample size – was entirely in Amicus’s control, and the company’s failure to power this group properly could cost it dearly. Approval in treatment-naive patients now looks extremely unlikely; even if the FDA does acquiesce, drumming up demand will be tough.

The regulatory chances for either of these two projects are uncertain, even before considering their commercial futures. Going in their favour is Lumizyme’s waning effectiveness, leaving longer-term Pompe patients in need of new options to keep the muscle-wasting condition at bay.

Sanofi provides a test case of sorts for regulators’ views on these mixed studies, with a May 18 Pdufa set for avalglucosidase alfa. Perhaps this time, Amicus investors will think harder before reading across from the outcome.