Welcome international healthcare professionals

Welcome, this website is intended for healthcare professionals with an interest in the ECC Webcasts. By clicking the button below you are declaring and confirming that you are a healthcare professional.

You are here

Leave No Stone Unturned: Assessing Novel Combinations with PI3K and CDK4/6 Inhibitors in HR+ AB

J. Cortés (Spain) & F. André (France)

SlotAA-4 Cortes, J.

Javier Cortés discusses the data from clinical trials with PI3K inhibitors and CDK inhibitors, looking at several compounds in clinical development and Phase I, II or III trials. The presentation considers future directions for the treatment of HR+ ABC.


I know that over the last year I think that we all agree that in patients with metastatic ER+ HER2- disease, endocrine therapy is the preferred option possibly.  If we do not need a quick response, if we are not in front of an urgent situation, this is quite clear.

But it is also true that usually there is acquired or the novel mechanism of different resistance and we have discussed a little bit some of them.  

Full transcription

Full transcription

I know that over the last year I think that we all agree that in patients with metastatic ER+ HER2- disease, endocrine therapy is the preferred option possibly.  If we do not need a quick response, if we are not in front of an urgent situation, this is quite clear.

But it is also true that usually there is acquired or the novel mechanism of different resistance and we have discussed a little bit some of them.  We have been discussing the mTOR pathway, we have been discussing the PI3K pathway, also the CDKs or whatever pathway, but also David discussed very nicely the HER2 pathway which at the end is one of the most important mechanisms resistant to endocrine therapy alone and we saw how trastuzumab or lapatinib might reverse the resistance with endocrine therapy.

Also we learned just three years ago that the mTOR inhibitor, sirolimus might overcome this mechanism of resistance of the PI3K mTOR activation of the signalling pathway and now we know that there might be other compounds which are in clinical development. 

We will have very early interesting clinical trials with PI3K inhibitors and also we have some clinical data also on more phase III trials with CDK inhibitors and this is what I would like to discuss with you, which is the clinical data we have today with PI3K inhibitors and/or with CDK inhibitors.


These are my disclosures.


And because we are going to discuss clinical activity about some drugs which have not been approved in Europe, I would like to make these disclaimers.

Several Novel Compounds Target the PI3K or Cyclin D-CDK4/6-Rb Pathways

It is difficult to talk about signalling pathways after Fabrice because he knows much more than me, so thank you so much because now it’s much easier.

But these are the compounds which are the majority of them are not all in the clinical development which we are going to discuss.  Why these and not others?  Because these are more ahead in the clinical development, that’s the reason and all of them have been evaluated or had been evaluated in randomised trials.  This is the reason we will discuss only these compounds.

Regarding the PI3K inhibitors, buparlisib and pictilisib block all the isoforms of PI3K.  Alpelisib and taselisib, alpelisib from Novartis and taselisib from Genentech Roche, seems to be specific against alpha or they are beta-sparing or whatever, but they are not against all the isoforms.

And regarding cyclin D kinases four and six we have these three compounds, palbo from Pfizer, ribociclib from Novartis and abemaciclib from Lilly.

Several Novel Compounds Target the PI3K or Cyclin D-CDK4/6-Rb Pathways

The question here is are these compounds able or will they be able to overcome resistance, again the novel or acquired resistance for patients with ER positive, HER2- breast cancer when we use endocrine therapy?

Several Novel Compounds Target the PI3K or Cyclin D-CDK4/6-Rb Pathways

Let’s start with the PI3K inhibitors. 

PI3K Inhibitors + Endocrine Therapy in HR+ BC: Preclinical Models

Do you know, if we administer PI3K inhibitors to a tumour cell which is ER+ it is very interesting to observe that the transcriptional activity might be increased because of an endocrine or an oestrogen receptor dependent mechanism of action.  This increase in the transcriptional activity might be blocked with the addition of fulvestrant, which is an endocrine therapy.

So you can see on this slide how if we combine alpelisib or buparlisip with fulvestrant we might induce much more activity, in fact they are synergistic against these tumour cells, against this xenograft compared when we use only the PI3K inhibitor or the endocrine therapy.

Phase I Trial of Buparlisib + Fulvestrant in HR+/HER2 - mBC

Which clinical data do we have?  We haven’t talked about many Phase I trials because these are the only data we have of some of these combinations, but just one Phase I trial per agent, no more than that.

Let me start with BKM120, buparlisib, the pan-PI3K inhibitor from Novartis.  I think it is important because at San Antonio this year we will have the first Phase III data coming from a PI3K inhibitor and this is it, so I think this is interesting to know.

Two or three points from here.  The first point; you can observe here that in this Phase I trial with 29 patients we observed responses. 

Phase I Trial of Buparlisib + Fulvestrant in HR+/HER2- mBC

We tested two different schedules, the continuous and the intermittent schedules and we observed responses with both types of schedule but the important point here is that the toxicity which was basically fatigue and transaminitis and rash was more important with the continuous schedule.  That’s why the preferred schedule was the intermittent dosing.

Phase I Trial of Alpelisib + Fulvestrant in ER+/HER2- ABC

What about BYL, alpelisib plus fulvestrant Phase I trial?  81/84 patients were analysed and there was something we observed in pre-clinical models.  If we administered this drug for patients with mutated tumours, the activity was in my opinion very interesting with about 25% of responses, but we do not have responses when these patients did not have mutations in the PI3K.

Phase I Trial of Alpelisib + Fulvestrant in ER+/HER2- ABC

I think that we have to look at this figure very carefully but it seems at least that the benefit in terms of progression-free survival seems to be – this is absolutely not a comparative trial, so this is just hypothesis-driven, it seems that that the median progression-free survival is very interesting for those patients with mutated tumours but it is not that interesting in patients with wild-type.

Alpelisib + Fulvestrant Had an Acceptable Safety Profile: Phase I Trial in ER+/HER2- ABC

What about toxicity?  I think this is absolutely not unexpected, so hyperglycaemia.  What is very typical from these compounds, 20% was easily managed with an anti-diabetic treatment and also unspecific maculopapular rash which is also a typical adverse event from these types of compounds.

Phase II Trial of Pictilisib + Fulvestrant in HR+ ABC Refractory to AI

Finally the only randomised Phase II trial with endocrine therapy and PI3K inhibitors we have today and this was the Ian Krop analysis presented in San Antonio last year.  You know that fulvestrant was compared versus fulvestrant plus pictilisib.  The trial was not positive for progression-free survival but there was a clear trend, so the sample size was small, the hazard ratio was 0.74, nevertheless.

However and of interest, there was absolutely no benefit in those patients with mutations compared to those patients without mutations so in this case at least for pictilisib it seems that mutations of the PI3K was not a biomarker to be tested.

I would like to highlight that the majority of the samples came from the primary tumours whic we know is maybe not the best way to do it, but these are the data we have.

Of interest, in those patients with ER+ and PR+ tumours, look at the hazard ratio i- 0.44 and the p-value was interesting, the improvement in median was interesting and that’s why there has been an amendment of this trial to include more patients with ER and ER+ tumours.

Regarding toxicity, there were no treatment-related deaths and the toxicity again was not unexpected with diarrhoea and rash being the most important Grade 3 and 4 adverse events.

PI3K Inhibitors and Endocrine Therapy Combinations in HR+ ABC

So what’s coming next?  Regarding buparlisib, BKM120 the pan-PI3K inhibitor by Novartis, we’ll have BELLE-2 and BELLE-3.  Again BELLE-2 positive results.  We have a press release saying that the trial met its primary endpoint.  It was positive for progression-free survival.  Let’s see the amount of difference and let’s see also biomarkers and so on, but we will have these data by San Antonio this year.

Belle-3 hopefully we will estimate to have some analysis by the middle of next year.

We make some comments about FERGI which was a negative trial for progression-free survival, but the hazard ratio was 0.74 and we also have SANDPIPER.  SANDPIPER is the randomised Phase III trial with beta-sparing like "quite similar to" alpha-specific in some way PI3K inhibitor by Genentech. This trial included patients who are progressing on aromatase inhibitors.

And finally solar-1 which is a similar trial but in this case with BYL which is the PI3K alpha-specific drug by Novartis.

Several Novel Compounds Target the PI3K or Cyclin D-CDK4/6-Rb Pathways

Let’s move to the cyclin D kinases four and six inhibitors, palbociclib, ribociclib and abemaciclib.

Several Novel Compounds Target the PI3K or Cyclin D-CDK4/6-Rb Pathways

The first thing is that these drugs are specific for CDK4 and 6 but it is not very active in blocking CDK1, 2 and 9 and I think that is something important to be considered afterwards.

Ribociclib + Endocrine Therapy in HR+ BC: Preclinical Models

These are preclinical data in animals showing tumour cells quite resistant to endocrine therapy. These tumour cells were treated with letrozole or ribociclib, the CDK4/6 inhibitor by Novartis and you can observe that these drugs did not work there but when you combine both they were synergistic and were able to stabilise or decrease the disease.

Phase Ib Trial of Ribociclib + Letrozole in ER+/HER2- ABC

Which data do we have in the clinic?  These are data coming from the Phase I trial of ribociclib and letrozole, so almost 30 patients were treated, 29.  One partial response was observed, 11 stable diseases and because there were many patients without measurable disease, that’s why we call something like ‘non complete response, non-progressive disease’. In my opinion this is like stable disease and that’s it, but you know, this is the definition.

Phase Ib Trial of Ribociclib + Letrozole in ER+/HER2- ABC

So what about toxicity?  Again it is very well known that this type of agents might produce neutropenia.  In this Phase I trial about 50% of patients did have Grade 3 and 4 neutropenia.  It was very well managed and reversible once treatment was interrupted.

PALOMA-1: Phase II Trial of First-line Palbociclib + Letrozole

What about palbociclib? With palbociclib we have more data, we have two randomised trials already presented, one of them published in the Lancet Oncology, PALOMA-1 which is a first-line letrozole with or without palbociclib in ER+/HER2- disease.  About 165 patients were treated and you know the results, neutropenia Grade 3, 4 in the range of 50-55% more or less but improvement in progression-free survival was amazing in my opinion.  The median was improved from ten months to 20 months and it was enough for the FDA to consider this approved based on an isolated approval basis.

PALOMA-3: Phase III Trial of Second-line Palbociclib + Fulvestrant

What about the most important randomised trial we have, because this is a randomised Phase III trial, a pivotal trial, it is the PALOMA-3 trial.  Basically it compares fulvestrant with or without palbociclib in patients resistant to aromatase inhibitors and the data again are amazing.

Very similar – 60-62% of Grade 3 and 4 neutropenia without any other very important toxicity in my opinion, 2% of fatigue but again the improvement in progression-free survival was amazing.  The hazard ratio was 0.42 more or less and the median went from 3.8 months with fulvestrant alone to more than nine months with fulvestrant plus palbociclib.  This data was published already in the New England Journal of Medicine by Nick Turner.

Phase Ib Study of Abemaciclib + Endocrine Therapy in HR+/HER2- mBC

What about the last compound, abemaciclib?  It also has been explored in a Phase I trial in combination with different options, with trastuzumab, with everolimus and exemestane and with different endocrine therapy, with fulvestrant, with anastrazole, with exemestane, with letrozole, I don’t know, with everything.

Which data do we have in terms of toxicity?  Diarrhoea - diarrhoea is the most important toxicity because of abemaciclib. 

Phase Ib Study of Abemaciclib + Endocrine Therapy in HR+/HER2- mBC

It seems to be much more important than neutropenia and here you have the data in terms of activity, so we observed responses when we combined abemaciclib with letrozole, with anastrazole, exemestane.  To be very honest, I don’t see very important differences between these three combinations.  I don’t know if it’s necessary, but nevertheless the data is there.

The overall response rate was in the range of 13% more or less but the disease control rate was in the range of 70-80% in total.

CDK4/6 Inhibitor and Endocrine Therapy Combinations in HR+ ABC: Phase III Clinical Development

So which data are upcoming?  Regarding palbociclib we have PALOMA-2, very early, which is a pivotal Phase III trial in the first-line.  We have PALOMA-4 which is similar but in the Asian population and we have the PEARL.  The PEARL I think is a very interesting randomised trial comparing palbociclib and exemestane compared to capecitabine, so I think that this trial is going to be interesting and important.  We have to watch to see where capecitabine can be used, when we can use also exemestane plus palbociclib.

What about ribociclib?  There are also three very interesting Phase III trials which are upcoming – MONALEESA-2 and MONALEESA-3.  MONALEESA-2 basically endocrine therapy letrozole with or without the cyclin D kinase inhibitor compound in the first-line.

In the first second-line with fulvestrant is MONALEESA-3 and also in premenopausal women which is one of the first trials which are going to specifically explore this situation, it is MONALEESA-7.

Finally with abemaciclib, MONARCH-2 and MONARCH-3 and this is vice versa so PALOMA-2 is similar to MONALEESA-2 and now similar to MONARCH-3, so MONARCH-3 is the first-line.

Now PALOMA-3 is similar to MONALEESA-3 and similar to MONARCH-2, so MONARCH-2 is in, I don’t know, resistant patients, so 2-2-3, 3-3-2, I don’t know, so I think this is something interesting so at the end we will have to -.  I will forget the names of these trials, to be honest, but nevertheless.

PALOMA-3 was positive, that’s all I know.

Future Directions for the Treatment of HR+ ABC

Now, what is coming next?  Two or three points, key messages which I think is important.  The first one is we need biomarkers, we need biomarkers so there is an important by the companies, by the investigators to try to demonstrate which biomarkers might be interesting to select which patients might benefit more or less.  This is something again in clinical development.

We also know we have to know which is the optimal sequence to use all these agents, so mTOR inhibitor followed by CDK inhibitor followed by PI3K.  PI3K first, CDK inhibitors first; we do not have many trials to answer this question but this is going to be very important in the very near future.

And finally because of the mechanism of resistance to all these agents, I that combining cyclin D kinase inhibitors with PI3K inhibitors and or with mTOR inhibitors makes sense and there are at least four clinical trials ongoing to try to define which is the optimal dose and how to move forward afterwards.


So just to conclude – we have one minute left.  Just to conclude, I think that we have several compounds which are in clinical development, some of them in Phase III trials or ones in Phase I and Phase II.

What is clear is that PI3K inhibitors and CDK inhibitors in combination with endocrine therapy seems to work more or less but there is activity there and the results from clinical trials support the use of these agents in endocrine positive HER2- population.

We now have the data of palbociclib in first-line and second-line showing that the amount of difference is tremendous and I think that in the very near future we need biomarkers and we need to know which is the optimal combination to move forward because at the end these patients will progress.

And that’s all – thank you so much.  [Applause]

Interim Assessment:  Poll and Panel Discussion

André:  Okay, thank you Javier.  We have the first question, or maybe some of you have some questions that you want to address, then we are going to go through all these questions.

The first question that I think is interesting to start is how do you see the future of treatment sequencing?  What would be the first-line, second-line or third-line?  Maybe Dr Cortés could give his opinion on that.

I would suggest to make a short answer because we have just ten minutes and several questions.

Cortés:  Today it’s very easy because the only approved drug is everolimus in terms of targeted therapy so first-line in my opinion is aromatase inhibitor alone, second-line exemestane plus everolimus, third-line fulvestrant.  We can discuss fulvestrant first or not, but it doesn’t matter, maybe it is not a topic for today.

In the very next future, assuming PALOMA-2 is as good as PALOMA-1 I think that first-line in my opinion could be CDK inhibitors in combination with aromatase inhibitors and in second-line I would wait to see the data of BELLE-2 to decide how to move forward.  I’m sorry, but although I know the data, I cannot say anything else at this time.

Cameron:  No, I think that’s absolutely right.  At the moment it’s fairly straightforward.  If palbociclib is approved in Europe it will become first-line and we need to know what is going to work after that and sitting in this room we don’t know the best sequence.

I suspect as we’ve done in endocrine therapy, we will learn over time how to sequence all these agents but right now we don’t have hard data to prove that one or other sequence is better after that.

André:  The second question is again about the sequence.  When PI3K inhibitor will become available, should we select them before everolimus or after everolimus?  I can take the question.

In fact there are different clinical trials.  There are some clinical trials testing the efficacy of PI3K after failure to mTOR inhibitor and there are some other clinical trials that include patients who have not been exposed to mTOR inhibitors so I think the results of these clinical trials will tell us what is the optimal setting for using a PI3K inhibitor.  I took the easiest question.

Cortés:  One very quick comment.  I think that BELLE-3 may also help us to decide that because BELLE-3 is on those patients who have had progression on everolimus.  That maybe might help also to decide how to move forward here.

André:  So here is a question for Dr Cortés.  Dr Cortés, are you available tonight for dinner?

Cortés:  It depends.  If the chromosomes are XX maybe; if XY, no.

André:  Okay, so then we have a question.  We were on the sequence, what is the optimal sequence and now we are going to move to the combination, a question related to combination.

The first question; should we consider combining mTOR inhibitor plus CDK4 inhibitor plus PI3K inhibitor and if yes, when?

Cameron:  I think we have no data.  There is one of the things you showed where there is a combination of exemestane, everolimus and one of the ‘ibs’ and I forget which one it was and I forget which of the CDK4/6 inhibitors – seven patients, one response so there is no suggestion from that study that a double combination plus endocrine therapy is suddenly going to be better.

For triplets, yes it may be fantastic but until we see clinical data, I wouldn’t do it.  It should be done in a trial.

André:  I think right now the effort on early phase development is to combine endocrine therapy plus CDK4 inhibitor plus PI3K or mTOR inhibitor, so this triplet, now people are learning how to combine optimally these three different triplets.

There are some other combinations also that are PI3K inhibitor and mTOR inhibitor; this is the combination between PI3K inhibitor to avoid the feedback route after mTOR inhibitor or to avoid the mechanism of resistance to PI3K inhibitors.

So there are some perspectives of combining all these agents but we need to know how to combine them.

Then again it is how to choose one versus the other one and the question is which biomarker should we look at if we have to choose between PI3K versus CDK4 inhibitor?  I don’t know, Javier if you want to -?

Cortés:  You know much more than me.  I think the first point here is to take into account that we have to analyse the metastatic sample or maybe liquid biopsies, but we have to forget analysing primary tumours – that’s my first point.

Let’s see the data, but if you have a mutation in PI3K it makes sense to block that pathway, but I really do not know because we do not have any data today, so it’s very difficult to answer that question.

If in the future we have biomarkers we might select which patients will benefit more from PI3K or whatever.  Maybe this is something we can discuss, but in my opinion today at least for me, I don’t know for Fabrice, but for me it’s very, very difficult to answer that question, I’m sorry.

Cameron:  I would agree and I think with a raft of different agents blocking these different pathways it is important that all of us, including the pharma companies, to invest in biomarker research so that given if they are licensed we learn the best way to use them for the right patients.  There is a risk that we will never know this if we don’t do the studies.

Cortés:  Nevertheless, everolimus-based therapy or palbociclib-based therapy, the hazard ratio in the range of 0.4 – 0.4, it’s a bit higher than trastuzumab-based therapy.  Trastuzumab hazard ratio was 0.62.  62 with trastuzumab, so now it is 0.4 without biomarkers, so I think that it is very difficult today to know how this field is going to move in the very next future.

André:  But probably because all these compounds are well characterised and the targets are well characterised so I think that as opposed to some other treatment like chemotherapy, I suspect we will be able to have some predictive biomarkers.  I think for PI3K the alpha specific, it seems that the mutations are quite predictive in the Phase I or early drug trial.

For the CDK4 inhibitor I suspect there will be some biomarkers in the next two or three years because the pathway is well characterised.

Then there is one question about whether oestrogen receptor positive advanced breast cancer or oestrogen receptor positive breast cancer could be the target of immunotherapeutics?

Right now there is not any public data on the efficacy of anti-PD1 in ER-positive breast cancer but there will be some data in the next three or six months.

There is one question I don’t understand.  It is any suggestion on adding trastuzumab in the subcutaneous formulation?  This is for our trastuzumab and T-DM1 colleague, Dr Cortés.

Cortés:  I don’t see that there is any point of adding trastuzumab, new formulation, subcutaneous trastuzumab?

André: I think maybe you can answer overall on the subcutaneous trastuzumab.

Cortés:  I think the data are there.  The HannaH trial seems to be very interesting, so subcu trastuzumab is not inferior to IV trastuzumab, so I think it seems to be more comfortable for patients.  In the peripheral trial patients preferred subcutaneous trastuzumab so why not?  I think that if possible that’s a very good approach.

André:  Then there is one question; will the mTOR pathway become the last line in fact?

I can take the question.  I don’t believe that.  The way things will be done is that when the patients will come with the metastatic breast cancer, we will identify whether this patient is more sensitive to mTOR inhibitor or more sensitive to PI3K inhibitor or more sensitive to CDK4 inhibitor and then we will select the first-line according to the prediction of sensitivity.

In the future, like in ten years from now I don’t think we will go sequence one then two then three.  We will select first-line, the drug for which the patient is the most sensitive.