PD1/ CTLA4 : Upfront or sequentially?

Bookmark and Share
Published: 17 May 2017
Views: 2599
Prof Christoph Höller - Medical University of Vienna, Vienna, Austria

Prof Hoeller speaks with ecancer at EADO 2017 about the timing of combination therapies using PD-1 and CTLA-4 targeted therapies to treat melanoma.

While these targets have been subject to a great deal of investigation independently, Prof Hoeller notes that there is no trial assessing the two in a direct comparison for upfront combination, or one after the other.

He considers the patient subgroups which may benefit most from these differing approaches, based on tumour infiltration and LDH levels, and weighs how toxicity may be managed.

Prof Hoeller also considers how best to assess patient responses to targeted therapies, based on expression of molecular markers including circulating tumour DNA.

Combining PD-1 and CTLA-4 was also discussed by Dr Christian Blank, here.

ecancer's filming has been kindly supported by Amgen through the ECMS Foundation. ecancer is editorially independent and there is no influence over content.

The two presentations I had at this year’s conference here in beautiful Athens at the EADO were, number one, to discuss if we should use PD-1 inhibitors in combination with CTLA-4 inhibitors up front or if it is a better idea to maybe sequence those two strategies and thereby maybe to have a longer duration of response with our patients. Looking through all the data that we have available on the topic, what I came up with in the end is that we don’t really have a clinical trial that actually up front compares those two strategies. So we just have circumstantial evidence and what the circumstantial evidence in the moment would suggest that there are some subgroups of patients, especially maybe patients where you are under pressure, where you need a quick response, a strong response, an immunotherapy, like patients with elevated LDH, patients with brain mets for example, that in these groups definitely these are the ones where all the data would implicate that you definitely will go up front with the combination.

But then there might be some groups like those really good patients that have low LDH, low tumour burden, that might have a high expression of PD-L1 or maybe, even better, a high infiltration of CD8 positive T-cells in the tumours. These are probably patients where we can achieve in the first line a very similar result, at least according to the now available data, with PD-1 monotherapy over the combination therapy. So these might be more the group where we might think about sequencing those strategies and going for one after the other. However, having said that, we don’t have a formal study that compares this and what we would really need is a prospective study that has one arm that has the combination immunotherapy, one arm that has the PD-1 up front then goes over to CTLA-4 upon progression and ideally a third arm that does not wait for progression but starts with one and then maybe adds on the other. I think this would be very interesting.

Was there any consideration of toxicity?

Definitely there is always consideration of toxicity because PD-1 monotherapy has significantly lower toxicity. However, one has to say that despite the fact that combination therapy is definitely more toxic also there the toxicity in experienced hands is manageable. So I would say that number one is always the consideration of efficacy and what I need in a specific situation and then clearly I have to factor in toxicity but knowing that if you follow the guidelines, if you have enough experience, you can manage all of those patients.

Are there any markers for a progression?

We don’t really have good markers for that beyond the radiological outcome. So if we see the progression happening on immunotherapy then that’s what we see. We might have some markers, and I’ve discussed some of these in my second presentation which was more focussed on targeted therapy but we can see the similar changes also in patients on immunotherapy where, for example, you could follow serum markers like tumour cell derived DNA and especially then you can follow up tumour specific mutation signatures in the patient serum. In some patients, using such a method, you can actually find out that these markers go up and would indicate progressive disease a bit earlier than your usual CAT scan staging. However, also with these markers you have some patients where it doesn’t happen, for example especially patients who develop brain metastases where it seems that these markers don’t go up as thoroughly as with other patients or as reliably. So you can add these on but you always have to consider that in the positive situation it will tell you something but in the negative situation you might have to also factor in other information.

Could you tell us about your second presentation?

The second presentation was the question on how we actually assess the response to targeted therapies because what we do nowadays in routine clinic is that we use just CAT scans every three months and then we factor in some serum based markers, mostly LDH levels and S100B, the melanoma tumour marker that we are using in daily routine. All of these factors clearly have some impact and we know that if LDH and S100 show a very nice drop-off during therapy that this would indicate a patient with a prolonged progression free survival. It’s not necessarily that these serum markers completely correlate with the RECIST criteria that we have in radiological assessments but, as mentioned, there are interesting new developments where you can actually monitor cell free DNA or tumour derived circulating DNA that you can just extract from a simple blood draw which clearly is an advantage because you don’t need to biopsy a tumour, you can just draw blood and do a liquid biopsy. As mentioned, it could be used to just look at the levels and see if this correlates with tumour regression or progression but what I thought was even more interesting is work that was published from the team around Richard Marais in Manchester where they actually can use these samples to detect upcoming new mutations, resistance mechanisms. So if this means that in the future we might have a very easy way to follow the development of resistance and maybe detect a resistance mechanism early, this could mean that it could help us to switch to a follow-up therapy definitely earlier and also not just to switch by the usual trial and error system but maybe have a guidance to which therapy would be now ideally fitted to that patient. So this is very fascinating work.

Then in imaging clearly everything in the moment, a lot of research is focussing on biological signals, on metabolic signals with the positron emission tomographies, with the different PET systems. There has been a lot of research on PET-CTs and we know that with targeted therapies we often see a very dramatic initial response in the PET signal and then if we see that very early on again it would indicate that this patient has a higher likelihood of having a longer response. But it is not always like a one-to-one relationship, it’s just an indication of a likelihood. New developments look now into coupling the PET method, or the PET system, with MRI instead of a CT scan which could actually give you better anatomic correlation, especially if you have metastases in the liver or in the brain, and also into new PET tracers that could actually maybe give you another level of information over the standard labelled glucose that we use at the moment.

So there are several of these new markers now incoming and they have to undergo very clear correlation with the things we use now. They have to show where they miss signals, where they have the strengths but I think they will be increasingly used in the clinic.