ASCO 2015

Pembrolizumab shows promising activity against head and neck cancer

Dr Tanguy Seiwert - University of Chicago, Chicago, USA

You were looking here at advanced squamous cell carcinoma of the head and neck. What was the context and what were you trying to do? What was the challenge?

These were recurrent metastatic disease patients with head and neck squamous cell tumours. They were heavily pre-treated, about 60% of patients had two or more lines of therapy so a very poor prognosis population. Generally we think of a median survival for this population of 6 or less months.

Now you’ve been trying an anti-PD1, a PD1 targeted therapy. If they didn’t get that what’s the standard treatment they might have got? Not too many options, I suppose.

Options are limited, we have chemotherapy. Usually in the first line setting for recurrent metastatic disease we use doublet chemotherapy with a platinum, either with 5FU or with a taxane. Sometimes we add cetuximab onto that, that’s the only targeted therapy. Cetuximab by itself has a moderate or small response rate of 10-13%. In the second line setting we may use methotrexate or a taxane and that’s really it. We have chemotherapy and we have cetuximab and so the options are limited.

Could you describe what you did here in this study?

In this study we used pembrolizumab, a new therapy, an anti-PD1 antibody, in patients that are heavily pre-treated with recurrent metastatic head and neck cancer. The dose was given as an infusion every three weeks at a flat dose of 200mg given every three weeks.

What happened?

The exciting thing about the treatment was that it actually worked quite well. One in four patients had a response, meaning substantial shrinkage. An additional 25% had stable disease and if you just look at any level of tumour shrinkage, so a very soft cut-off of benefit, even more, 57% of patients, had some level of tumour shrinkage.

It sounds pretty amazing. What did you make of it?

I think these are remarkable data, something that we’ve actually not seen before. If you compare this with cetuximab, the best targeted therapy we have right now, cetuximab has a response rate of 10-13%, this is twice as good, it has double the response rate.

What was the PD1 directed agent doing to these patients?

It seems like cancer in general and head and neck cancer in particular, these cancers have found a way to escape the immune system. The immune system can fight the cancer. What pembrolizumab or PD1 inhibitors do is they make the cancer visible again to the immune system. So rather than treating the cancer directly, we are empowering the immune system to fight the cancer. So what you sometimes see early on is that these cancers become inflamed and they get infiltrated with these immune cells and the immune cells actually do the job. Sometimes I compare this to lighting a fire and that fire keeps burning. Because it’s not directly a treatment for the cancer but an indirect treatment the side effects are also oftentimes much less.

Other agents?

There’s another PD1 agent called nivolumab that’s been developed for head and neck cancer and there’s a PDL1 agent called MEDI4736 that’s being developed for head and neck cancer. There are additional agents in development although I’m not aware of head and neck specific trials right now.

Some of the standard therapies for this group of patients are quite toxic. What about these new agents?

Chemotherapy is toxic, radiation is toxic, cetuximab is better tolerated. But these agents are remarkable because they have a very mild side effect profile. The biggest side effect or the most common side effect was actually mild fatigue, grade 1 or grade 2. Generally speaking, compared to what we currently use, these are very well tolerated agents. Rarely there are more severe side effects we have to watch out for those but that’s just really just one or two cases in our 132 patient cohort which is a very substantial cohort.

Some of your patients would have had HPV involvement and others not. Did that make a difference?

It’s a great question. About 60% of the patients were HPV negative, 40% were HPV positive, so we had a nice balance. It was active in both subgroups, that’s again remarkable because there is now data suggesting that cetuximab, another therapy for EGFR, doesn’t work so well in HPV positive. So this seems to work in both entities.

And you got responses, were these durable responses?

That’s the other really exciting thing about pembrolizumab in head and neck cancer, 86% of patients who responded continued to be in response. So it seems like we have durable responses, something that we don’t typically see with chemotherapy or with targeted therapies.

Of course predicting the outcome of PD1 therapy is an interesting thing and you’ve got a genetic marker, haven’t you?

That is correct. There are two things that I think right now may help us identify those patients that benefit. What has been widely studied is PDL1 immunohistochemistry and that does a reasonably good job. But the new thing that we are showing here at this meeting is that there is also a gene expression signature that seems to be, at least in these early data, exceptionally good at predicting which patients should not receive the treatment and is still quite good at predicting who should get it. So I think if this is validated this might really help us identify the right patients.

And that feature is?

It’s called an interferon gamma signature. There’s very likely a big role for immunotherapy with PD1 agents like pembrolizumab in head and neck cancer. I think we have to await more mature data, we have to await survival data. My expectation is that they likely will have a significant impact on survival so I think it’s very likely oncologists will in the future be able to hang their hat on this type of therapy. I think right now it’s important to shunt people to clinical trials to really obtain mature data and do it safely.

Other PD1 agents?

There is another PD1 agent that’s being developed in head and neck cancer, that’s nivolumab. That’s also being tested in head and neck cancer.

So promising. What do you think are the clinical implications of some of these developments?

For head and neck cancer I think it provides a lot of hope. Right now I think it’s very reasonable to refer patients for immunotherapy trials because I think there is a signal that these agents might really help patients. I also think the larger context is that very likely we’ll see a change in how we treat cancer and we’re starting obviously with the very worst patient population, patients who are heavily pre-treated, but I foresee that there’s a good chance that this will move to front line or even the curative setting. I think it will change head and neck cancer treatment and, even broader, it will change many cancer types.

So in just a few words what should doctors take home from this?

Pembrolizumab is active in head and neck cancer. One in four patients respond and the side effect profile is very tolerable.