Event – A defining moment for IDO inhibition
There was a time when a wave of novel immuno-oncology targets looked capable of repeating the success of PD-(L)1 and ushering in the next stage of the anticancer revolution. Just a few years later only one such target – IDO – has come close to delivering on the promise.
Much depends on a study billed as the biggest binary event of 2018: the Echo-301 melanoma trial of Incyte’s IDO inhibitor epacadostat, combined with Merck & Co’s Keytruda, which should read out mid-year. Not only will this be the first pivotal readout for the IDO class, it is expected by some to answer once and for all the question of whether there is substance behind the IDO hype.
The question is relevant not only for Incyte, whose IDO is its most important asset after the marketed drug Jakafi, but also for other players that had piled into IDO inhibition, including Bristol-Myers Squibb, Roche, Newlink, Astrazeneca, Pfizer and Nektar.
|% of market cap||48%|
|Event||Data from phase III Echo-301 trial|
IDO is an enzyme that breaks down tryptophan, prompting T cells to become inactive – a mechanism that appears to be involved in materno-fetal tolerance, and one that tumour cells can hijack to damp down immune response.
The big question that Echo-301 must answer, therefore, is to what extent blocking IDO extends the benefit already seen with anti-PD-1s like Keytruda. The key measure to look for when Echo-301 reads out is median progression-free survival, as the data might be insufficiently mature when the study is unblinded for the other co-primary, median overall survival, to have been reached.
The Keynote-006 trial suggested that Keytruda monotherapy’s PFS benefit in first-line melanoma was 5.6 months at the best dose. In Bristol-Myers Squibb’s Checkmate-067 trial, meanwhile, adding Yervoy to Opdivo gave median PFS of 11.7 months – though at the expense of serious adverse events.
Thus it would seem that if in Echo-301 epacadostat can add a similar survival duration to PD-1 inhibition as Yervoy does, but without a sharp rise in toxicity, this would be a win. Indeed, a recent Credit Suisse survey found that most investors expected Incyte’s study to yield median PFS of 12 months or more.
Investors trying to guess the outcome have one study to go on: Echo-202, an uncontrolled, phase II trial of Keytruda plus epacadostat in various tumour types.
Last year’s Esmo meeting saw data on a melanoma cohort from this study, with first-line patients showing median PFS of an impressive 22.8 months. The number is likely to fall with longer follow-up, and the result in a blinded study will naturally be worse than in an uncontrolled one like Echo-202; still, the signs are good.
|Comparison of selected trials in first-line melanoma|
|Keytruda||Opdivo||Opdivo + Yervoy||Keytruda + epacadostat|
|mPFS||5.6 mth||6.9 mth||11.7 mth||22.8 mth*|
|Grade 3-4 AEs||16%||55%||20%**|
|Note: *this number could be lower with longer follow-up; **discussant's slide cited 26%, but Incyte has pointed out that this was a mistake, since it double-counted some patients. Source: Esmo 2017.|
Echo-301 completed enrolment of 706 melanoma subjects last September. The participants had not received prior systemic therapy except Braf/Mek inhibitors, if relevant, meaning that effectively they were first line; they will certainly all have been anti-PD-(L)1 drug naive.
As this is an event-driven, double-blind study it is impossible to determine precise timing of data readout. Incyte has refused to disclose how many “events” – deaths – will trigger study unblinding, but has confirmed that the primary analysis will relate to all-comers; subjects were stratified by PD-L1 expression, however.
While Echo-301 is largely seen as a huge binary outcome, Michael Gladstone, a partner at Atlas Venture, recently wrote that the readout could answer more subtle questions, too.
These include whether PD-(L)1-naive patients were the right population to treat in this setting, and whether IDO-selective inhibition is the best way to hit the tryptophan metabolism pathway. Investors will also be trying to gauge epacadostat’s potential in other settings – Incyte has hinted at adjuvant melanoma, and a large programme is under way in lung cancer.
Competitors will be waiting with bated breath, too. Merck is one of several companies, including Astrazeneca and Bristol-Myers Squibb, with non-exclusive epacadostat alliances, but serious money will only flow from exclusive deals, like the takeovers of Iomet by Merck and Flexus by Bristol.
Those last two deals set up the possibility of commercialising wholly owned anti-PD-1/IDO combos, whose relevance should grow on a positive Echo-301 readout. Incyte has set up an in-house scenario of its own, licensing in Macrogenics’ anti-PD-1 MAb MGA012 last year.
Indeed, given the money that has been invested in this field it might be that developers refuse to give up on IDO even if Echo-301 reads out negatively. After all, there is always a way to put positive spin on ambiguous data.
|Pivotal trials of anti-PD-1 + IDO inhibitor combinations in melanoma|
|Study||Company||Anti-PD-1||IDO inhibitor||n||Primary endpoints||Trial ID||Primary completion|
|Echo-301||Incyte||Keytruda||Epacadostat||706||PFS, OS||NCT02752074||May 2018|
|NLG2107||Newlink||Keytruda or Opdivo||Indoximod||624||Safety, PFS, OS||NCT03301636||Dec 2019|
|–||Bristol-Myers Squibb||Opdivo||BMS-986205||700||PFS||NCT03329846||Aug 2020|