Max Nisen is a Bloomberg Gadfly columnist covering biotech, pharma and health care. He previously wrote about management and corporate strategy for Quartz and Business Insider.

The market for a new class of immune-boosting cancer drugs may be the most closely watched in the pharma world. It's also one of the most unsteady.

The fortunes of five mammoth companies with FDA-approved drugs in this category -- Bristol-Myers Squibb Co., Merck & Co., AstraZeneca PLC, Pfizer Inc. and Roche Holding AG -- depend in various degrees on how this market shakes out. 

Last year, Roche's entry in the sweepstakes, Tecentriq, got what the FDA calls an "accelerated approval" to treat previously treated bladder cancer. Permanent approval depended on a new trial confirming the encouraging results of an earlier trial. But Roche on Wednesday announced that confirmatory trial had flopped, throwing the immune-oncology (IO) market into fresh turmoil. 

The FDA may pull the drug's approval for this subset of patients, and the drug's broader efficacy is now in question. Investors should take this as another forceful reminder that this new class of drugs is prone to cycles of disruption and hype and should be treated with caution. 

Brakes
Roche shares fell after its cancer drug Tecentriq failed a late stage clinical trial
Source: Bloomberg

If the FDA retracts Roche's accelerated approval, then that will be bad news for a drug that has gotten off to an impressive start. Tecentriq managed more than $100 million in sales in the first quarter, only its fourth quarter on the market. An expected European approval in previously treated bladder cancer patients is now likely off the table.  

And this news raises fears that other medicines may fail trials designed to confirm promising early results. Three other drugs have gotten accelerated approvals to treat bladder cancer, with two of those approvals coming in the past 10 days. The accelerated-approval pathway is one reason these drugs have expanded to so many cancers so rapidly. Roche's failure makes it a bit harder to count on the ever-escalating sales ramp-up analysts expect for these medicines.  

Blue-Sky Thinking
A Roche trial failure is a reminder of the risks of analyst assumptions of skyrocketing immune-oncology drug sales
Source: Bloomberg

Much depends on what exactly is behind Roche's failure.

This is the second time another company has failed in an area where Merck's drug Keytruda has succeeded. Merck is awaiting an FDA decision on its own successful bladder-cancer trial. Keytruda also succeeded in treating newly diagnosed lung-cancer patients, where Bristol's drug Opdivo failed

There's a small chance Merck's drug is superior to at least Opdivo and Tecentriq. That might lead oncologists to prefer Merck's drug in cancers where several of these drugs compete. 

But Merck has also restricted many of its trials to patients most likely to benefit, while other firms have targeted broader populations, in hopes of expanding the potential market for these medicines. Roche's stumble might simply be due to overly aggressive trial design -- or just bad luck.

If so, then Roche's failure may be isolated. But there's also a chance these drugs have less upside than companies have hoped, and that the FDA should demand more evidence of their benefit before granting wide approval for drugs that cost more than $12,000 a month. 

These new concerns come atop a bevy of other uncertainties for the sector. Firms are testing dozens of different combinations of these medicines with other drugs and treatments. One or more of these approaches could dramatically reshape the market in the coming years. And we've yet to see how much having three or more potentially substitutable drugs in a given cancer will impact pricing power. 

Today wasn't the first time an IO drug has had its prospects unexpectedly upended in a single day, and likely won't be the last.  

Update: An earlier version of this column transposed IO drug sales estimates in the "Blue-Sky Thinking" chart. 

This column does not necessarily reflect the opinion of Bloomberg LP and its owners.

To contact the author of this story:
Max Nisen in New York at mnisen@bloomberg.net

To contact the editor responsible for this story:
Mark Gongloff at mgongloff1@bloomberg.net