Let's Talk About Met
March 3, 2014 was a sad day – it was when we announced that we’d stopped the Phase III METLung study of onartuzumab due to a “lack of meaningful clinical efficacy.” Onartuzumab, also known as MetMAb, was designed to block the MET protein, which is overproduced in half of cases of non-small cell lung cancer (NSCLC).
Many people at Genentech, including myself, believed that onartuzumab was potentially a real breakthrough for lung cancer. The METLung results weren’t just disappointing for us. They were disappointing for the people affected by lung cancer – the cancer that claims the most lives in the United States.
It's a disease where new medicines are clearly – desperately – needed, and one that carries a heavy stigma that can affect treatment decisions. Read Race, Wealth, Gender, Cancer? for more on that topic.
The speed at which the METLung study enrolled reflected how excited the lung cancer community was about onartuzumab – and likely how disheartened they were with the results.
In METLung, we tested a targeted combination and applied advances – like companion diagnostics – to select patients most likely to respond.
So what happened? And what have we learned?
MONDAY MORNING QUARTERBACKING
What was surprising about the METLung results was that a personalized, biomarker-guided approach didn’t work.
We quickly moved onartuzumab into a Phase III study based on promising Phase II data that showed a survival benefit with onartuzumab in people who had high levels of MET on their tumors. We followed the science and our theory about targeting MET. The pieces seemed to fit.
But was there a clue in the Phase II results that suggested the survival data was not as promising as it appeared?
Looking back, we didn’t see a significant difference in tumor shrinkage (response rate). Although this is not unusual, in retrospect, we still might have questioned more why people were apparently living longer if their tumors weren’t shrinking.
A NEGATIVE ISN'T ALWAYS A FAILURE
Even though the ultimate outcome of METLung was negative, it doesn’t mean it was a failed trial. It means we didn’t get the answer we were looking for.
This is the nature of science. We have to follow what the data tell us. In this case, the data does not support continued development of onartuzumab for lung cancer. However, the story doesn’t end here.
It takes courage and commitment for a patient to participate a clinical trial. And we owe it to those who participated to learn as much as we can in order to move research forward.
We will apply what we learn from onartuzumab to everything we do – from selecting targets to further developing our pipeline of investigational lung cancer medicines.
Here are a few ways we are carrying the lessons of METLung forward:
Don't Fish for Red Herrings
A mountain of research suggested that MET was an important culprit in cancer. But maybe we followed the wrong lead. Maybe an overabundance of MET is a symptom of another problem.
The METLung data suggest that cancers may not be “addicted” to the MET pathway in the same way they are to other mutations. Looking at this more closely may help inform how we choose targets moving forward.
MET may in fact be very important to human cancers, but maybe it enlists other proteins that do the dirty work to make tumors grow. Perhaps we need to block other proteins that interact with MET, or target other pathways that talk to the MET pathway, like the RAS-RAF and PI3K pathways. You can read more about following the PI3K pathway from my colleague, Lori Friedman.
Diagnose the Diagnostic
Crucial to the METLung study was a diagnostic test assessing the amount of MET present on lung tumors. This test was based on immunohistochemistry (IHC), a complex technique that requires a trained pathologist to complete and interpret the results.
Were the results interpreted consistently? Was the test sensitive enough? Are there other types of tests we could have used instead? Answering these questions may give us further insight into the METLung results.
WHERE WE GO FROM HERE
Ultimately, there’s no way to avoid uncertainty, ambiguity and complexity in the research and development of new medicines.
That’s why we’ll keep tackling the complexity in lung cancer. We’re pinpointing new targets, exploring new diagnostics and developing new medicines. Lung cancer is the leading cause of cancer death worldwide, and we won’t rest until lung cancer is a thing of the past.
True failure is to ignore what the evidence tells us, or to learn nothing from the answers we don’t want.
The only path forward is to stick to the principles of good science. Ask the right questions. Design the right experiments. Assess the evidence honestly.
And when you think you're finished, ask more questions.