Rounds with the Investigators 2012 | Lung
QUESTION: My patient is a 40-year-old woman with a strong family history for early cancer deaths, presented January 2011 with numbness on the right side of her face. MRI of brain showed 17 ring-enhancing lesions. PET/CT scan revealed 5-cm hypermetabolic left upper lobe lung mass with supraclavicular, mediastinal and bilateral hilar lesions along with liver lesions and axial skeletal disease. Biopsy showed adenocarcinoma, and she was treated with whole brain irradiation in February followed by pemetrexed/cisplatin x 2 cycles, then switched because of toxicities to carboplatin/paclitaxel, with bevacizumab added around the fourth cycle. She did not tolerate any of these treatments well and sought a second opinion with me about 3 months into treatment, and I requested testing of her initial tumor sample for EGFR and ALK mutation. This came back positive for EGFR, and I started the patient on erlotinib in July of 2011. She tolerated erlotinib well. Surveillance PET and MRI brain in October showed “dramatic response.” In March of 2012, she complained of worsening weakness and bone pains and was found to have a new 1-cm lesion in her brain and clear progression of her known bony disease. Last week, I sent her for stereotactic radiosurgery of the brain and palliative radiation for the bony lesions. In this patient who has been poorly tolerant of standard chemo and bevacizumab, progressed on erlotinib, what is the best treatment option at this point?
DR ERIK RUPARD: A 40-year-old woman with a strong family history for cancer. She had a mother, father and a sister, all of whom had metastatic disease of unknown primary early on and probably didn’t seek much medical treatment.
So she initially presented in January of 2011 with numbness on the right side of her face. It was going down her right arm. Ended up getting an MRI in the emergency room, which showed — I counted them — 17 ring-enhancing lesions, including her brain stem and her cerebellum and probably quite a few that were below the limits of detection.
So the subsequent workup included a PET scan, which showed a 5-cm hypermetabolic left upper lobe lung mass with a bunch of supraclavicular mediastinal bilateral hilar lesions along with, unfortunately, multiple liver lesions, axial skeletal disease, and almost all of this was symptomatic. She had pain all over.
So she eventually got a biopsy of a supraclavicular lymph node, which was consistent with adenocarcinoma and was initially treated with whole brain irradiation in February of 2011, followed —
DR NEIL LOVE: Did she have mutation testing?
DR RUPARD: She did not, initially. Yes. This was at another facility. I inherited her later, but no, she did not at that point in time.
She switched to carboplatin/paclitaxel because of toxicities. I guess she was in that group that Dr Patel mentioned that didn’t tolerate the pemetrexed very well. And to be honest with you, she didn’t really tolerate anything. She initially got — so she got the pemetrexed/cisplatin followed by carboplatin/paclitaxel, and then around the third or fourth cycle they added bevacizumab.
DR LOVE: So just relevant to this question of need for CME — and I don’t know what your perception was in terms of what was going on in the other hospital. But Tom, we’re talking about February 2011, a year ago, a 40-year-old woman who’s a nonsmoker who hasn’t had any mutation testing.
DR THOMAS LYNCH: I would have liked to have seen mutation testing at that point. I mean, I think that I would, if I knew she was an EGFR mutant, I probably would start with erlotinib and probably not with whole brain radiation, even with 17 lesions. But again, it’s hard to go back and change.
DR LOVE: Just to pick up on that, Jyoti, do you agree in terms of withholding whole brain radiation in a patient who has an EGFR mutation?
DR JYOTI PATEL: We generally do.
DR RUPARD: Really? Even if somebody was symptomatic like her?
DR LYNCH: Yes.
DR RUPARD: Hmm.
DR LYNCH: Because this is exactly the patient who could live 5 to 6 to 7 years. And 5 to 6 to 7 years after whole brain —
DR RUPARD: I see.
DR LYNCH: Now, granted, she’s 39, so she probably doesn’t have very much vascular disease yet. And she’s probably not going to get dementia, but early in the whole course of the EGFR story we treated people with whole brain radiation then gave them erlotinib or gefitinib. And the number of patients I had who had substantial cognitive decline was very large, particularly in older patients who — because most of our metastatic lung cancer patients don’t live 5 to 6 years afterwards to have the effects of whole brain.
DR RUPARD: So she was very symptomatic from the standpoint of her CNS lesions. How long does it take, as opposed to whole brain irradiation, for the erlotinib to help with those symptoms?
DR LYNCH: So, by saying that — I don’t mean to be cavalier — I think I would start steroids. I would give erlotinib at the same time. I would follow her like a hawk. In some patients you’re not going to get away with it. Okay? In some patients you will have to give radiation, either stereotactic or whole brain radiation. So I’m not trying to say that you can do this for everybody. But I at least would try my best to see if we could get away with that. There are several examples of very nice responses in the CNS to primary erlotinib-based therapy.
DR LOVE: Just out of curiosity, I mean, how would you compare the responses that you see in the CNS to responses that you see in people with EGFR mutations with systemic disease?
DR LYNCH: I think it’s very similar.
DR LOVE: Jyoti?
DR PATEL: And I would agree with that. We’ve even seen patients with great responses who present with leptomeningeal disease, which is always so difficult. But one thing, to concur with what Tom was saying, the rapidity of response with erlotinib is amazing, like 5 to 7 days, which is exactly what you would imagine for radiation. So it’s not that you’re putting — you’re waiting too long.
DR RUPARD: Okay.
DR LOVE: Steve?
DR STEPHEN GRABELSKY: Would you use the standard-dose erlotinib? Or would you do the higher-dose pulse erlotinib that has been talked about for patients with the CNS metastases?
DR LYNCH: It’s a great question. I would use the standard-dose erlotinib. I was at a fantastic seminar this week by Jeff Engelman from MGH, who talked about how we should be rethinking our use of kinase inhibitors in general. And we really should be thinking about pulse dosing as a way of taking advantage of signaling technology and what’s happening inside the cell. But in this setting, I would go with the standard dose. But I’d keep an open mind. Maybe a couple of years from now we’re going to be going to more pulse doses of kinase inhibitors.
DR LOVE: Just out of curiosity, I always thought the pulse dose thing was about trying to get a higher dose so you get more drug in the CNS. Is there any reason to think with systemic disease the pulse dosing makes sense?
DR LYNCH: Just that you may allow cells to go through apoptosis better and there may be a rationale for giving pulse dose as a way of — and you may actually, paradoxically, delay the emergence of resistance. It’s not — this is theoretical. It’s not proven yet.
DR LOVE: I want to actually bring this case up to date and get more input, but just one other follow-up question in terms of this issue of rapidity of response: We talked about the fact that we like to poll investigators. And one of the things that we just did a poll on, Tom was talking about, was the question of the patient who let’s say you know has an EGFR mutation but is very symptomatic. They have extensive lung disease. They’re short of breath. Jyoti, is there any reason — in breast cancer, sometimes that kind of situation will get people to use chemo before hormones because of this thinking it works faster. What about in terms of these patients?
DR PATEL: I think for patients who are EGFR mutated, erlotinib is faster than chemotherapy and they are more likely to respond. You want to give them the best option, particularly when you’re meeting them — as a lung cancer oncologist, you sound a little bit ridiculous, but it — particularly when you’re meeting them in the intensive care unit and talking to the pulmonologist about how you’re going to get the erlotinib down.
DR LOVE: Wow! Okay. Erik, can you bring us up to date?
DR RUPARD: Sure. So, after not tolerating the treatments very well but having some clinical response, her boyfriend, now her husband, brought her in to see me for a second opinion. And I called the pathologist and requested retrospective testing of her sample for EGFR and ALK, and it came back, not surprisingly, positive for EGFR. So I started her on erlotinib in July. And within a couple of months she’d had a dramatic response, further dramatic clinical response and, in addition, she was tolerating the therapy better. I did a surveillance PET scan, an MRI in October, so about 3 months after starting therapy, and the radiologist just called me personally and asked me if it was the same patient. It was one of those situations.
So she had a surveillance scan that showed a really nice response. Then, unfortunately, in March of this year she started to complain of worsening weakness throughout her body and bone pains. And this had been coming up in January, February, and we adjusted some pain medicine that she had been on, and I felt like some of it was related to that. But by March it became clear to me that something else was going on and it was time for her surveillance scans.
So, in addition, she had this tremor. And she’d always had this right-sided tremor ever since her disease was diagnosed. It had gotten much better when I started the erlotinib. And now it was starting to come back, visibly. And this is a tremor in the right side of her lips and the right side of her hand, going all the way down. So I repeated her MRI, and she was found to have a new 1-cm lesion in her brain. And I also did a bone scan on her, because she was having this all-over body pain that was not severe but certainly progressive. And, unfortunately, she appeared to have substantial progression of her known bony disease.
So this is now 9 months into erlotinib therapy, and she is unfortunately progressing through erlotinib. So my question is: Where do you go from here?
DR LOVE: Tom, a common question?
DR LYNCH: So I would definitely give stereotactic radiosurgery to the primary brain lesion. I think that’s the easy part.
DR RUPARD: Yes, but she’s received that already.
DR LYNCH: So that’s the easy part. How many bone mets are we talking about, more than 2 or 3 that are growing?
DR RUPARD: Yes. They aren’t even discrete bone mets.
DR LYNCH: Okay.
DR RUPARD: They’re — it’s just diffuse —
DR LYNCH: Diffuse.
DR RUPARD: — axial skeletal disease. Correct.
DR LYNCH: So in this setting I think your best chance is to do everything you can to get her on a study of cetuximab and afatinib. So afatinib is a dual kinase inhibitor that inhibits EGFR and HER2. It’s an irreversible inhibitor. It’s a drug which we should have some data coming out this year that looks at it in first-line setting of lung cancer compared to chemotherapy, a study from Asia that we’re awaiting the results of. There are data now that show that afatinib plus cetuximab is our best second-line option, regardless of what the resistance is due to, whether resistance is due to T790M or not due to T790M.
What I would like to do in this setting, if possible, I would love to biopsy this person, because you could learn something from the biopsy that would change your opinion. For example, this person has a 15% chance of having small cell as their histology. If they had small cell, you would eagerly give the person cis or carbo plus etoposide, depending upon your favorite regimen for small cell.
DR LOVE: Could you just elaborate a little bit more on this small cell transformation and where you see it?
DR LYNCH: So about 15% of EGFR mutants, when they progress, if you biopsy, you can find small cell lung cancer phenotype. They still maintain the EGFR mutation. The driving mutation is still there when you look inside the small cell, but they can respond to chemotherapy for small cell. So that’s one of the reasons why I think that has importance.
The other thing is: You may find MET amplification. If you found MET amplification, you may go down a different road toward treating the relapse, and you might want to consider using a MET inhibitor or looking for a trial that has a MET inhibitor as an option at that point.
If, let’s say, you said, “The patient says, ‘No way do I want a clinical trial. I refuse to do a clinical trial,’” what would I do next? Depending upon the amount of progression, if you’re telling me there are 20 or 30 spots that are growing, I would probably stop the erlotinib and add chemo. If you told me there were 3 to 5 spots that were growing, I would probably continue the erlotinib and add chemotherapy at that point. And it would be whatever your favorite second-line chemotherapy is. You’ve already given her carbo/pem. Carbo/paclitaxel, you’re probably looking at something with gemcitabine.
DR RUPARD: Gemcitabine.
DR LYNCH: Yes.
DR LOVE: Although she didn’t do well on chemotherapy. You mentioned trying to get her in a trial of afatinib/cetuximab. Like, where would he look? Or could you get a patient in a trial like that?
DR LYNCH: So our — actually, we’ve filled ours up right now, so that the issue is I would go to ClinicalTrials.gov. We actually had an open study with afatinib and cetuximab that was open, which closed about a month ago. And there are 1 or 2 trials that are open still with that combination. Cetuximab, obviously, is commercially available. Afatinib, our hope is it will get approved in first line, and that will allow us an off-label use in this setting. But again, we don’t know the results of the trial. We don’t know if it will be approved in first line. There’s a lot that’s unknown at this point. But it’s by far the most exciting regimen. Vince Miller and Greg Riley, the group at Memorial, along with a number of other people, published on this. It’s a very exciting new regimen.
DR LOVE: Just to clarify, though, what about — I mean, theoretically, you could do erlotinib and cetuximab.
DR LYNCH: There’s really no evidence that erlotinib and cetuximab would have benefit in this setting, so I would not do erlotinib and cetuximab in this setting. This is all based on some terrific mouse modeling work that William Pao, who’s now at Vanderbilt, and Katie Politi, who’s now at Yale, did when they were post docs for Harold Varmus in Hal Varmus’ lab, where they did some really exciting mouse modeling, suggesting that this combination was able to overcome acquired resistance. And initially, they thought it was just T790M that benefited. Now we know that T790M is interesting, but the non-T790M patients seem to respond just as well as the T790M patients did.
DR LOVE: Do we have any update? The only thing I’ve seen was at last ASCO where they presented that.
DR LYNCH: Yes.
DR LOVE: Excuse me. Is there any other group that’s looked at it? Was there any follow-up to that trial? And also, that trial, you see numbers, incredible waterfall plot. Most of these patients responded. But did you have any feeling or, even from your own practice, what it meant clinically? Were these useful responses?
DR LYNCH: So I don’t have any additional data to share. I do know that the responses are the kinds of responses — they’re 6- to 9-month responses, and then you develop resistance at that point. So they’re not cures in that setting.
But what’s interesting in a lot of these cases is that you can also go back to the well, sometimes. So you can use afatinib/cetuximab for a while. Maybe you find a small cell phenotype when you biopsy again. You can go back and consider reusing erlotinib down the road. So I think we’re going to see a fair amount of rotation of therapies and using repeat biopsies to drive treatment. I think that’s very important in EGFR-mutated cancers, is using repeat biopsies to drive the treatment. And I believe that it’s clinically indicated to do that now, not just for research.
DR LOVE: Just to clarify a little bit more, too, when you hear erlotinib/cetuximab, you think about theoretically pretty bad dermatologic possibilities. What about afatinib/cetuximab?
DR LYNCH: So that’s a great point. Mario Lacouture, the dermatologist on — I think, Neil, you’ve worked with Mario —
DR LOVE: Sure.
DR LYNCH: — in programs before. It can cause skin toxicities and GI/bowel toxicities. It’s not as easy a regimen as erlotinib alone, obviously. And it’s something that you want to be watching very carefully with your patients.
DR LOVE: So, just to follow up, Jyoti, how do you approach, in general, patients who’ve had EGFR mutations, great responses, and now progression?
DR PATEL: I’d echo a lot of what Tom says. I think one thing to understand is patients with T790 mutations can also do quite well. Greg Oxner has published survival curves that look better from the start or be — the onset of resistance, these patients do better than your average patient with non-small cell lung cancer.
Our group with Melissa Johnson is working with Greg Riley doing a Phase I trial looking at an HSP90 inhibitor, AUY-922, in combination with erlotinib. And that’s a chaperone protein. So you restore that genotypic driver with erlotinib. So far we haven’t seen a lot of toxicity. We haven’t seen a lot of great responses, but it may be that, again, if you have a small amount of response, to restore stable disease or durability of response might be okay.
In general, if these patients have just oligometastatic progression, so progression in one area in the skeleton, I’ll just give radiation and continue the erlotinib. And I, too, will often interchange. If the patients are tolerating erlotinib and they have multiple sites of systemic progression, I’ll probably do single-agent therapy and continue erlotinib and drop it to 100 or 75 and then give them 4 cycles of chemotherapy and get back to the erlotinib, because there is selection of different cells. In patients who I need to hit a little harder with doublet chemotherapy, usually stop the erlotinib and then go back on erlotinib as maintenance.
DR LOVE: So I’ll open it up for questions in a second, but one more question about this case back to you, Tom. What about this interesting family history this lady has? She has a mother, died age 64 with mets to the brain. Do we know what the primary was from that, Erik?
DR RUPARD: We don’t know. No. Unfortunately.
DR LOVE: And then father, 58; sister, 45, mets, unknown primary. Tom, any associations, family history-wise, with EGFR-mutant lung cancer?
DR LYNCH: There’ve been a couple of families. Two families, I believe, who have had the T790M as a germline mutation. But I’m not aware of any other familial EGFR-mutant syndromes in lung cancer. The challenging thing in lung cancer, looking at heredity and family history, is that it’s very, very difficult to separate smoking habits and smoking patterns, which tend to run in families from other factors in terms of family history.
DR LUIS RAEZ: I have a quick question. Now when patients have an EGFR mutation, we put them on erlotinib. They fail. Now the biopsy report, the second one, half of the time comes with 790. Will you continue with erlotinib in these people, or you only continue if you get the other 50% that you don’t know if it’s a 790?
DR PATEL: So I, if it’s T790, I actually continue the erlotinib by and large, and then —
DR RAEZ: And then you add —
DR PATEL: — look for a trial. Right. Right. We’ve picked up a couple of patients with small cell transformation. They act differently. They have more rapid progression of disease. And for those patients I stop the erlotinib completely and do chemotherapy.
One question to ask: Did you treat with a bisphosphonate or denosumab?
DR RUPARD: Yes. Since the day she walked into my door, she’s been getting zoledronic acid. Yes.
DR LOVE: Jyoti, any thoughts about bone-targeted therapy in lung cancer? And what about denosumab?
DR PATEL: So I usually give bone-targeted therapy when my fellow tells me to. I usually forget, because, one, most of our patients with non-small cell lung cancer I don’t think live 3 or 4 or 5 years to suffer a pathologic fracture. So, by and large, unless — if someone has asymptomatic, low-volume disease, I treat the cancer and do chemotherapy alone.
For patients who have painful bony metastasis or who have known bone lesions in the femur or in the pelvis, sometimes I’ll go ahead and start — I am still using primarily bisphosphonates in patients who have good creatinine clearance. I tend to use denosumab in patients who’ve failed that indicator. Again, for our patients I don’t think one is better than another. And our patients are getting infusional therapy, by and large, anyway, so adding the other infusion compared to the subcutaneous injection, I don’t think changes their lives.
DR LOVE: Maggie, just out of curiosity, right now, how do you approach bone-targeted therapy, metastatic lung? But particularly metastatic breast, are you using a lot of denosumab?
DR MARGARET DEUTSCH: I am not using a lot of denosumab except for patients with renal insufficiency. I expect that I’ll probably use — I mean, there is a convenience factor. I mean, if I have somebody who has bone-only disease and they’re getting fulvestrant, then it’s easy to give them a sub-Q injection and not giving the long infusion. But if they’re already getting infusional chemotherapy, like you, I tend to just give them another infusion.
DR LOVE: Tom, how do you approach bone-targeted therapy, and are there patients with bone mets, but, again, because of the natural history of the disease, where you just don’t do it?
DR LYNCH: I think Jyoti’s completely correct. I think that for those patients who’ve got a high burden of bone disease, I do think zoledronic acid is a very reasonable option. I have not been using denosumab, only because I haven’t been using it. I can’t tell you that that’s right or wrong. It just hasn’t become my personal habit yet. And I do think that it’s a bigger issue in someone like your patient with an EGFR mutation who may live 3 to 4 years to develop bone fractures.