ECC 2015

Atezolizumab for locally advanced or metastatic PD-L1-selected NSCLC

Dr Benjamin Besse - Institut Gustave Roussy, Villejuif, France

Benjamin, it’s great to see you here and tell me about your work, where you work and what you’re doing at that lab because you’re now doing really interesting work in checkpoint inhibition.

Thank you for having me. I’m a medical oncologist in Gustave Roussy in Villejuif, France. I’m the Head of the Thoracic Unit, so I’m dealing mostly with thoracic malignancies.

Now you are one of the investigators contributing patients to the BIRCH trial. What is the trial doing?

It’s a trial, I have presented the results today, so it’s a phase II trial, single arm, of atezolizumab in patients enriched with PDL1. Meaning it’s a trial designed for PD-L1 positive NSCLC patients. So atezolizumab was given as first line in one cohort, second line in the second cohort and in the third cohort third line plus. So it’s the only trial that really tested the same drug in different lines in 667 patients, so it’s a very large trial.

It is a big study and the drug is directed against PD-L1 so having a PD-L1 rich population should help. What did you find?

We have screened more than 3,000 tumours to find that PD-L1 was overexpressed in 34%. So when these patients were treated with atezolizumab the response rate was quite similar across the three cohorts, meaning in first line, second line or third line plus, with 17% response rate in second and third line plus, 19% in first line. What we have shown is that when you restrict the analysis to the patient with very high positive NSCLC, it’s roughly half of the patients included in the trial, then you increase the response rate to almost 25%.

So there’s a sensitivity to the presence of this biomarker?

Clearly, so where to enrich? It’s not on/off, when you are a high expresser you do not respond all the time but it’s a good way to enrich a population so also to offer the treatment, for example, in first line.

This is, therefore, a clear clinical signal but is that signal strong enough to put it into clinical practice?

We need randomised data. So there was the POPLAR data that were presented, so it’s a trial of atezolizumab against docetaxel but it was a phase II trial. The phase III trial in an unselected population second line is ongoing. So this will be the data that we will need to put the drug into clinical practice. It’s currently tested up-front, so in first line setting, either with chemotherapy in unselected patients, either in the PD-L1 positive NSCLC patients compared to chemotherapy, so single agent compared to chemotherapy, but in this selected population.

How do much do you guess is the response better in the selected population, ones with the biomarker, as compared with the unselected population?

We will have this data with the randomised trial in second line but what POPLAR told us is that when the immunohistochemistry is negative, meaning the tumour is completely PD-L1 negative, it seems that the drug is no better than chemotherapy. But we need more data to know really how to use this drug, data in non-squamous, squamous, because the biology is a bit different. We need more data.

Everything is a trade-off, what toxicities do you get?

It’s what you expect with any PD-L1 inhibitor or PD-1 inhibitor. So the grade 3/4 adverse events related to atezolizumab was 11% so it’s lower than what you get with any chemotherapy, The toxicity profile is a bit different than for chemotherapy but still it was very well tolerated - a bit of nausea, but roughly all the side effects were below 10% any grade and very few grade 3/4 side effects.

OK then, in summary, what could be the clinical implications of this sort of finding that’s emerging at phase II if it goes through and fulfils itself further down the line?

We have already on the market a PD-1 inhibitor. It’s approved in squamous, it will be probably approved in non-squamous in second line. It might be that’s a competitor of this drug because it’s developed in the same setting. For the first line we have to see the data. Almost all the drugs have some development in first line, we will see what is the more clever development or the more accurate selection of the patients based on the PD-L1 positivity.

And a quick take-home message for cancer doctors right now?

These drugs, the checkpoint inhibitors, are standard of care as second line treatment in squamous. How we’ll have to use it in second line non-squamous, this is a bit of debate now but these drugs are in the clinic and when they work they can really work for a long time which is quite the difference with chemotherapy. You may have very, very long responders so it’s something very new in the metastatic lung cancer field.

Thank you very much, Benjamin.

You’re welcome.