DR SOCINSKI: So this is frustrating. It’s like the old sports — Wide World of Sports, the thrill of victory. You diagnose a sensitivity mutation and then the agony of defeat. Six to 12 months later, you’re dealing with progression.
There are lots of issues. We would rebiopsy to kind of look at what the mechanism of resistance may be. Fifty percent of these people have the T790M mutation. There are some directions to go in from that point of view. There are other mechanisms of resistance, MET amplification, that may take you down a different road.
There’s some encouraging data with afatinib/cetuximab. There’s some encouraging data with the Clovis compound, but we need a good strategy for the dominant mechanism of resistance, which is the T790M. There are issues of do you continue the TKI. Remember, these patients tend to be more sensitive to platinum-based therapy. Let’s say you rebiopsy and you really don’t find anything at that point. She’s progressing. One option would be to give her chemotherapy, a platinum-based doublet plus or minus bevacizumab as an option. What do you do with the TKI at that point? Controversial. Many would continue it with treatment. Many would interrupt and maybe go back to it afterwards. We have some trials ongoing addressing these sorts of things.
DR LOVE: So, knowing that there’s not one perfect option, if you had to say in general what you think you would do, what specifically would you do?
DR SOCINSKI: Assuming that she didn’t have anything that would lead me to a clinical trial, I probably would treat her with carbo/pemetrexed. I’d look at her brain, make sure she didn’t have any brain mets. I would consider using bevacizumab in this setting. And I would give her a trial of 4 cycles of that and then reassess her at that point.
DR LOVE: And as you started this chemotherapy — with bevacizumab, incidentally?
DR SOCINSKI: If she didn’t have any contraindications. Yes.
DR LOVE: Right. Okay. So as you started this treatment, would you be continuing the erlotinib or not?
DR SOCINSKI: I don’t. I say we’re going to put a hold on the erlotinib. We’re very likely going to go back to it, if you respond, as maintenance. But I don’t think it’s necessarily wrong to continue it, but what you’re doing is just — even though I think we can give these TKIs with chemo, the more drugs you give, the more toxicity you have. And if I’m not sure there’s a clear benefit, I’m probably going to interrupt and then go back to it afterwards.
DR LOVE: So, again, the art of oncology. And what people want to know is: What do you do? So I’m going to make it real quick around the table. Greg, I’m hearing carbo/pem/bev, if she’s eligible, and hold the TKI. Is that what you would do or not?
DR RIELY: I think one question I’d ask: How many liver metastases?
DR BELANI: Many.
DR RIELY: Many. Okay. So my key distinction being, if there’s one liver metastasis, everything else is controlled, I would strongly consider a local therapy option, radiofrequency ablation, something like that, something that can buy us time ‘til we have to make the decision.
DR LOVE: She also progressed elsewhere, correct?
DR BELANI: Yes.
DR LOVE: Yes. Okay.
DR BELANI: She progressed in the lung.
DR LOVE: Okay. So she’s got plenty of liver mets.
DR RIELY: But she’s got broad progression of disease. I would generally continue erlotinib and add chemotherapy, carboplatin/pemetrexed/bevacizumab is a perfectly reasonable combination to add, so that would be reasonable.
DR GOVINDAN: I would keep erlotinib. I would add pemetrexed and carboplatin.
DR LOVE: And just out of curiosity, what would you be thinking about maintenance as we head in that direction?
DR GOVINDAN: So I would — if she does extremely well, I would leave her on pemetrexed and erlotinib.
DR LOVE: So I didn’t ask you two about maintenance. How would you approach that?
DR RIELY: I would continue the erlotinib/pemetrexed/bevacizumab.
DR SOCINSKI: I would — for maintenance, if she had a response, I would do bevacizumab/erlotinib.
DR LOVE: Okay. Dave?
DR SPIGEL: Yes. A little different. I would not continue erlotinib. And I would start carbo/pem. I would do that for 4 cycles. And I’d continue pem alone.
DR LOVE: John?
DR HEYMACH: Yes. I would do carboplatin/pemetrexed and bevacizumab. And the reason, I think something that’s sometimes appreciated, it seems like EGFR mutants are actually more sensitive to VEGF blockade than others. If you look at the BeTa study that was comparing erlotinib and bevacizumab versus erlotinib, the group that really benefited from adding bevacizumab was actually the EGFR mutants. And there’s other data supporting that. And preclinically, we see that VEGF gets upregulated by EGFR. So in the EGFR mutants, whenever I can, I try to include bevacizumab. So I would do carboplatin/pemetrexed/bevacizumab. In the maintenance setting based on the ATLAS findings, I would go back to erlotinib with bevacizumab.
DR RIELY: We initially reported when patients came off EGFR tyrosine kinase inhibitors that they had a flare reaction. And when we defined “flare,” we said that patients who ended up in the hospital or died, that would be considered a flare. And we took all of our patients who went on clinical trials with next-generation EGFR inhibitors, where they had to stop erlotinib for the trial during the washout period, and it looked at their outcomes.
And what we saw is that about a quarter of patients during that washout period ended up in the hospital or dying. So that was concerning for us and so, as a consequence, we thought the idea really is that you should be continuing erlotinib when we start our next line of therapy. And to build off of that, Dr Halmos did a randomized trial. This was designed several years ago, and so the idea was to look — this was before routine use of erlotinib in the first line. And so as a consequence — and this was also designed before pemetrexed was used as a first-line agent. In this trial, patients with acquired resistance to erlotinib were randomized to either chemotherapy, pemetrexed or docetaxel, versus chemotherapy with erlotinib. And that was — it’s a relatively small study, and it was actually completed early, is my understanding, due to slow enrollment. They had 46 patients in total, comparing 24 in the chemo-only versus 22 in the chemo plus erlotinib.
And in the end, they saw no benefit for continuation of erlotinib. And there was some additional toxicity, of course, with the continued use of erlotinib in terms of rash.
DR LOVE: Sounds like pretty tiny numbers.
DR RIELY: Yes. I think that’s the big concern when you are doing a small trial, or when you’re doing a trial. You end it early for reasons that are practical rather than theoretical, and then you’re unable to really look at the endpoint in the way you wanted to. I think it was a reasonably designed trial, but stopping it early made it hard. There was no real suggestion, though, from the data that there was an improvement for the combination. So I think that gave them some support in the idea of saying no difference, because the hazard ratio was actually worse for the combination arm.
DR LOVE: And can you put this in context with other data, clinical data, addressing the same question?
DR RIELY: It’s challenging. For as long as we’ve been asking the question — and I know that you’ve been asking the question of us for a long time — this is really the only prospective data that we have that targets this question. And it’s frustrating that it’s a relatively small study and it closed early. So we don’t have — I don’t think this is a definitive answer. I think this is a question we still need a prospective trial to answer.
DR LOVE: So it seems like there’s also some mixed feelings about what to do clinically here. Ram?
DR GOVINDAN: So I just wanted to add a few things. EGFR mutation is actually present almost always in the founder clones. In other words, this is a causative lesion, causing transformation. This is not quite later in evolution of the lung cancer. So in other words, this is what causes the disease in these patients.
DR LOVE: Hold on. How do you figure that out, and what’s a founder gene?
DR GOVINDAN: So when we do this whole genome sequencing, we can actually do this, especially deep digital sequencing, we can look for the variant allele frequency. So this is actually fairly standard stuff now in genomic studies. So if you have these tumors, when you look at all this mutation prevalence, you can look for the population of cells that are clustered together with the variant allele frequencies. And EGFR mutation will always be the highest number expressed in the very early cells that have evolved. And this is the concept of clonal evolution. So EGFR mutation is something that starts off. And so in every subpopulation, you will find EGFR-mutant clones.
DR LOVE: Same thing with ALK?
DR GOVINDAN: ALK, we don’t know. We haven’t studied ALK to the extent — I can talk about EGFR confidently. But most importantly, when we biopsy these patients who have been on EGFR TK inhibitors, who are already EGFR mutations, we have never seen the sensitive mutation disappear. In other words, the exon 19 deletion will always be there. It was there, if it was there before. And what happened is subclones evolve, but the original clone that had the EGFR mutation never disappears. And so that’s why we have this problem. If you don’t treat them, that clone will expand, and that’s where you see in studies when there is a treatment break from EGFR TK inhibitors, when you start them back again, you see some tumor shrinkage.
Because the sensitive clone never goes away, it’s just been overrun by other clones, people like us tend to keep them on erlotinib. I don’t think that this is something that we can say, “You have to stop,” or, “You have to continue,” because what happens is if you don’t continue erlotinib when they have disease progression, at some point, if you’ve given them a break after 3 or 4 or 5 months, you can go back and re-treat them. You will see some decrease in tumor size. That’s because now the sensitive clone has expanded even more, because it’s biologically a more advantageous growth pattern. So that’s why we will never settle this issue by doing the studies, biologically.
Like the progression-free survival improves with EGFR TK inhibitors in the front line, but you never improve overall survival. So you can either give it at the time of disease progression, continue it, or you can do it much later after a break. So I don’t think there’s one way or the other we can say this is the right thing to do.