ImmunoGen sucks up the cost of novelty value
This article was originally published in Scrip
The clinical disappointment for Hoffman-La Roche's Kadcyla (trastuzumab emtansine) in December 2014 has forced Boston-based ImmunoGen, which licenses the toxin and linker of Kadcyla to Roche, into an expensive $200m deal on its near-term royalties on Kadcyla for cash to maintain R&D spending. Under the deal, ImmunoGen will receive the cash in early April 2015 from TPG Special Situations Partners (TSSP), a specialist finance house, and then will see nothing of the Kadcyla royalty stream until TSSP has received at least $235m (if the royalties flow in) or $260m (if it's a trickle).
That's a 17.5% or 30% premium on the advance, and that is just the start of the costs. ImmunoGen has also agreed that when the $200m is paid off, it will continue to pay TSSP 15% of Kadcyla royalties as long as Roche continues paying them to ImmunoGen.
ImmunoGen's immediate problem started in December 2014 when the top-line results of Roche's Phase III MARIANNE study showed that Kadcyla on its own or with Roche's other Her2-targeted drug Perjeta (pertuzumab) was as good, but no better than Herceptin (trastuzumab) in treating first line advanced HER2-positive breast cancer. That meant, in effect, that ImmunoGen was contributing only novelty to Kadcyla, an opportunity for Roche to extend its Her-2 franchise beyond the expiry of Herceptin's key US patent in 2019, but one that would be limited by the emergence of biosimilar versions of Herceptin.
In response to Roche's December news, stock analysts reduced their projection for Kadcyla peak sales from $3bn a year to around $1bn, also cutting ImmunoGen's slice from around $150m a year to $50m. And that meant that the company would not be able to fund its own ambitious $100m-plus a year development program from its income.
ImmunoGen's broader underlying problem is overstretching. Having been surfing a wave of enthusiasm for antibody-directed toxin conjugates that has seen its toxin-linker systems deployed by a wide range of pharma companies including Amgen, Bayer Healthcare, Novartis, Sanofi, Lilly and Takeda, as well as Roche, the company wanted to get out from under its dependency on other firms and their development programs. It has four of its own compounds in development, the most advanced of which is IMGN853, a folate-receptor targeted ADC, due to enter Phase II studies in endometrial and/or ovarian cancer in 2015.
However, shifting from out-licensing technology to in-house drug development is expensive. Since Kadcyla was initially approved by the US FDA in February 2013, ImmunoGen has raised its R&D spending from $69m in 2012 to $107m in 2014. In addition to Kadcyla royalties, the company also has received some $39m in 2014 income from licensing fees and milestone payments. But it still registered a $71m loss in 2014. With just over $107m in the bank at the end of January, the company either had to curtail R&D spending or raise more money.
The company would have found it difficult to return to the public markets with a secondary offering: its last public offering, in July 2012, raised $96m net of expenses at $16 per share: the company's share had been trading at $17.5 in February 2014 but had weakened throughout the year with the Kadcyla setback sending the share price down from around $10 in early December to $6 after the announcement. It has recovered to $8.5 since then.