ASCO 2010 Annual Meeting, 4—8 June 2010, Chicago
Interview with Professor Jean-Charles Soria (Institut Gustave Roussy, Paris)
Systemic therapy – what lies ahead?
Thank you very much indeed for coming to talk to ecancer. You’re very busy at ASCO - you’re an invited speaker all the way from Gustave Roussy in Paris; what are you talking about?
Well today I’m the invited speaker in a session dealing with personalised medicine in locally advanced, non-small cell lung cancer. So patients who are not resectable nor metastatic and have these huge masses in the lung and this is an area where we haven’t done much progress yet. One of the issues and paths of improvement was the idea to combine radiotherapy with targeted agents, but the reality is that until now this has been a very frustrating field: in one case we combined antiandrogenic agents like Avastin with radiotherapy and we increased tumour sensitivity, we increased tissue sensitivity and we ended up with major fistulas in the trachea. The other one was the combination of EGFR tyrosine kinase inhibitors with radiotherapy and we ended up with an arm which was the experimental arm below the standard arm.
So probably the big issue in locally advanced non-small cell lung cancer is to personalise therapy. And that comes with the capacity to predict whether the most important issue is local recurrence or metastatic recurrence. So you pick up where you’re going to put the weight on therapy – rather radiotherapy, rather systemic therapy. And to do so it’s very clear we need to do more molecular portraits of the tumours of these patients. There is not a single signature, transcription signature, done in stage 3B patients able to predict whether you’re going to recur locally or metastatically.
Second, obviously and yesterday a plenary presentation on this new compound called Crizotinib targeting a subset of lung cancer patients, 5% of them having a specific translocation between ML4 and ALK, is opening the way for a new era in lung cancer where you have objective response rates close to 65 -70%. So EGFR mutated patients, they got Iressa, they got Tarceva; the ML4 ALK, they got Crizotinib and of course the HER2 mutated ones, they already have Herceptin. So this disease which seems to be a large disease is being sliced into different parts like a salami with these slices sometimes having very specific compounds able to have a substantial benefit and major tumour shrinkage.
So you routinely, at Gustave Roussy, do genomics on your tumour samples. How small a sample can you work with and what’s the best way to get it in this kind of patient that you’re talking about?
I think the key issue is how many tumour cells you get in the sample and obviously the standard bronchoscopic technique sometimes gets a small number of soft cells and necrotic cells. So when bronchoscopies are done we need really to have at least two or three pieces. The alternative to that is obviously doing a CT-guide biopsy and you need to use a widebore needle to get that and that’s enough to get DNA. What we do mostly is not a full transcriptional profile because the lung is not a disease in which that is so helpful at this time. The most helpful things are to look for specific mutations so we do specific PCR screening for mutations.
But you’re not short of specific mutations in non-small cell lung cancers, especially in the smokers. ALK is particularly relevant to the non-smoking adenocarcinomas.
You are correct.
Most of your patients will have a hundred mutations, won’t they? The problem is which ones are driving the process and which ones are the passers-by.
You are right. The problem is very different between the never smokers, but the never smokers are increasing, you know - up to 20% of the patients who come for consultation at the Institut Gustave Roussy. But the smokers, indeed, is a much more complex situation. Nevertheless, when you have a smoker with the PI3K mutation this is really pinpointing a specific way we should be going. There is new data apparently about FGFR amplications in this subset of patients. So I think we’re going to start having some new agents even in the heavy smoking population.
That’s very exciting. So you’re doing a whole lot of phase I studies and phase 2 and randomised phase 2 and so on, where are you going to go after this ASCO? Are you going home, are you going to put more accent on this area or on that area? Is mTOR going one way, MEK inhibition another way? What are your personal ideas for the next year or two?
I think there are two issues here. The first one is that we have seen in this ASCO that the combination of targetted] agents are bringing something. There is very impressive data in breast with the combination of an mTOR inhibitor and an IGF1R antibody. What we have also seen unfortunately is that, contrary to what we were hoping, a combination of targeted agents is extremely toxic; there have been very clear presentations showing that in fact combining two targeted agents that seem to be very friendly end up in additive toxicity. So this is going to be a challenge but it’s one of the paths forward.
The second path, I think, is we really need to start thinking differently in phase I. We cannot keep in this process where we take all comers, three plus three escalating dose without molecular profiling the patient and enriching the population. I think when we do that retrospectively we don’t do it right because we use tumour samples that are all of three or four years’ disease when we are treating something totally different. Who believes that the molecular portrait of a resected colorectal cancer is the reflection of the disease of a metastatic colorectal cancer that went through FOLFOX, Oxaliplatin, Avastin, Cetuximab, Vectibix? It’s nonsense, we need to start doing molecular portraits on the true issue that is the reflection of the tumour you are treating. When you marry someone you look at her, you don’t look at the old picture how now she was.
If you look at the mother, of course, that’s always a very…
That’s very important.
Jean-Charles, thank you so much for coming and joining ecancer.