There are now three approved drugs that target PD-1 either directly or indirectly, and more are under development.While these drugs all seek to inhibit PD-1’s immune suppressive activity, they are quite different in the ways that companion diagnostic tests play a role in their use.

Nivolumab (OPDIVO®) – This antibody from Bristol Myers Squibb targets PD-1 directly, and has received approvals for multiple indications. It is already a $2 billion drug for BMS, and is leading the market for PD-1 inhibitors. Nivolumab is interesting in that BMS is heavily marketing the fact that a biomarker test is not required for the drug. This is an interesting spin in an era when most companies are promoting the precision medicines in their portfolios, but it may be contributing to the drug’s market leading position if BMS’s aggressive marketing is any indication.

Pembrolizumab (KEYTRUDA®) – This antibody from Merck also targets PD-1 directly, but its label specifies the drugs for patients who are over-expressing PD-L1 as identified by a companion diagnostic test. Given that both drugs have the same mechanism of action, this differentiation is surprising. Pembrolizumab is also generating just a quarter of the sales of nivolumab.  I haven’t had a chance to compare efficacy data, but if they are similar then it looks like Merck may have made a strategic error in utilizing PD-L1 as an inclusion criterion for its clinical trials if that was an elective decision.

Atezolizumab (TECENTRIQ®) – This antibody was developed by Genentech and is the newest PD-1 inhibitor on the market. Atezolizumab actually inhibits PD-1’s action indirectly by binding to PD-L1. This approach is considered superior by some because it allows PD-1 to function normally in some situations. Atezolizumab is in a strange middle ground between its two competitors with respect to companion diagnostic testing. While its label doesn’t require (or even recommend) testing for PD-L1 expression, it does state that patients with higher PD-L1 expression had a threefold higher response rate. Roche also introduced a new “complementary diagnostic” test for PD-L1 along with atezolizumab, and in the press release stated that it can be used by physicians to help identify patients who may respond to the drug.

I won’t delve into the continuing ambiguity around FDA’s concept of a “complementary diagnostic” test (which this most recent release does nothing to clear up), as the topic has been covered elsewhere in this blog (here and here). The key question that remains regarding the role of companion diagnostic testing and PD-1 inibition is whether PD-L1 expression or over-expression is important for efficacy.

The patient outcome data will ultimately tell the story, and should also indicate whether one inhibitory approach is more effective than the other. In the meantime, with what seems to be limited guidance from FDA, drug developers will have to determine for themselves how best to go to market.