It's a little sad that you can get more cancer information from the business websites than from the medical ones, but if you checked a story on Forbes.com today you learned that Bristol-Myers Squibb (BMS) provided a press release that one of their important Erbitux (cetuximab) trials didn't meet its primary endpoint of improved progression-free survival for chemo with Erbitux compared with the same chemo alone.

Erbitux is another inhibitor of the epidermal growth factor receptor (EGFR), similar to Iressa and Tarceva, but unlike those oral pills, Erbitux is an IV drug that is actually a monoclonal antibody to the part of the receptor that is on the outside portion of the cancer cell (extracellular). Erbitux definitely has activity in some cancer types: it's FDA-approved in treating colon cancer and head and neck cancer. But there have been some negative studies with Erbitux as well in other tumor types, including a large trial of chemo with or without Erbitux in pancreatic cancer that showed no benefit to the Erbitux combination (abstract here). It's also been studied in lung cancer, primarily in NSCLC, with some modestly encouraging results, but definitely not a slam dunk. I've described some of this work in a prior post.

Whether Erbitux is going to have any real role in NSCLC is a question that has fallen on a couple of pivotal randomized trials. The trial that we heard a bit about today was called BMS CA255-099 (they aren't always this catchy!), which enrolled over 600 patients with previously untreated advanced NSCLC to receive chemo with carboplatin and either taxol or taxotere, and half of the patients also received weekly Erbitux with that. Today, BMS explained that the trial did not meet its primary endpoint of improved progression-free survival for the people who received chemo with Erbitux compared with chemo alone (as assessed by an independent radiology review committee). However, this same committee, as well as the investigators involved with the study, did find that the people who received the chemo/Erbitux combination did have a significantly higher response rate.

Where does this leave us? I don't think it's good for the prospects of Erbitux, but with a significantly higher response rate, the door is still open for subsequent trial results. I have argued here at OncTalk that survival trumps response, whether we're talking about chemo or EGFR-based therapies like Tarceva, or Erbitux. I'd consider it highly unlikely that the overall survival of the Erbitux recipients will be better than the chemo alone arm if they didn't have a better progression-free survival -- subsequent treatments make it only more likely that the two groups will only converge over time. On the other hand, the rigid requirements of our new systems for grading response and progression can cause us to mask a lot of information in this classification system. It's also very important to note that progression-free survival is a very tough endpoint to show a difference in, because you're generally looking for progression every 2 or 3 cycles, so it's not a variable that can be as all over the map as overall survival, which isn't limited to something happening only when you happen to look. So although I don't think this makes me any more excited about Erbitux in NSCLC, I don't think this tells the whole story by a long shot.

There are certainly a couple of big trials out there still waiting to tell us something. One is called the FLEX trial, in first-line advanced NSCLC and done in Europe, comparing cisplatin/navelbine alone to the same chemo with weekly Erbitux:

(Click to enlarge)

Imclone, the company that developed Erbitux, is also running its own registration trial (designed to submit for approval by the FDA and commercial use) in the second-line setting, in which patients are randomized to chemo with Taxotere or Alimta alone (both FDA approved as second-line chemo options and shown to produce very comparable results) or one of these agents with Erbitux:

We should get information from these trials in the next year or two, and they should clarify whether Erbitux has something to offer in advanced NSCLC. It's worth noting that while these trials have often required patients to have tumors with EGFR protein expression by immunohistochemistry (IHC), it's possible that only a subset of patients are going to benefit from Erbitux. Next I'll talk about some evidence suggesting that certain patients on these trials may benefit more and others less with Erbitux.