Here is the continuation of my conversation with Dr. Nasser Hanna, lung cancer expert at Indiana University and all-around good guy (not part of his official title). Here we discuss a patient of mine who combines the challenges of managing stage IIIA N2 NSCLC with the issues of how to potentially integrate an EGFR inhibitor for an Asian never-smoker, particularly in a setting where we don't yet have good data.

At the tail end of the case, we also take a bit of a detour in discussing the question of whether the order of therapy matters as much as getting in the right therapies over time, specifically with regard to trying to prioritize chemo vs. an EGFR inhibitor.

====================================================

Dr. West: This case is of a 54 year old lifelong never smoking Asian woman who was involved in a motor vehicle accident and went to the emergency room and she had a chest x-ray there followed immediately by a CT of the neck and chest. She had a 2.4 centimeter right upper lobe spiculated mass and she also had several lucencies in her C2 verterbral body, very non-specific appearing. She had musculokeletal injuries in her sternum and actually a pelvic fracture.

She subsequently had a PET scan that showed the right upper lobe mass had an SUV of 4.5. Other findings were consistent with trauma, nothing suggestive of hiler or mediastinal uptake, nor any distant disease.

She had a CT guided biopsy that showed a moderately differentiated adenocarcinoma. The spine MRI was equivocal, and she actually proceeded to an open biopsy at C2 that was negative for metastatic disease. She had a mediastinoscopy that revealed multiple non-enlarged nodes in several different stations on the right positive for involvement of up to 8-9 mm in each, and nothing on the left.

At this point she has multi-station, microscopic N2 disease that did not appear on the PET scan. What approach would you be considering here? First of all, would you be inclined to send off for EGFR mutation testing and perhaps use that in this setting where it's not stage IV, but rather stage III disease as far as we can see? And if you would pursue a more standard chemo/radiation approach, would it be with a thought to moving on to surgery, or would you rather pursue chemo/radiation with a curative plan but no anticipated surgery?

Dr. Hanna: Jack, that is an incredibly difficult case.

Dr. West: That's why I'm sending it your way.

Dr. Hanna: Her performance status prior to the motor vehicle accident was good?

Dr. West: Excellent.

Dr. Hanna: And she had not been losing weight and this was just found incidentally because of her motor vehicle accident...

I think that because she has multi level N2 disease, that indicates to us that her chance for a cure is substantially lower, and the likelihood that she has systemic disease is very high. In fact, we know if we do local therapies only, which means surgery only or radiation only, in someone who has multi-level N2 disease, even if it's just microscopic, that their 5 year survival is somewhere between 10 and 15 percent. That would suggest that 85-90 percent of these patients do have systemic disease.

So my bias would be to give her preoperative therapy, and at Indiana University we would give her two courses of cisplatin and etoposide with concurrent radiation to 45 Gray. We would repeat the PET scan about 4 weeks after her completion of treatment, and assuming that she had not progressed or developed new disease somewhere else, we would offer her a lobectomy with a complete mediastinal lymph node dissection.

At this point I would do no further therapy. Now the question about EGFR mutation testing is a very provocative one. If you were to do that testing, it would imply one of two things. Either that you would intervene therapeutically with erlotinib now, or you would intervene therapeutically with erlotinib if she were to develop metastatic disease.

Let's take the second issue first. You would assume that her metastatic disease resembles her primary disease if you were to check EGFR mutation testing now. That's a bit of an assumption. It could be that here metastatic disease is EGFR wild type. Maybe that's the bad disease. So I would be reluctant to use that information to treat her with erlotinib in the future. I would want to know what her current status is, particularly if she recurred say one or two years later. But you cannot assume it's the same disease.

Much like in breast cancer, you can't assume that because the initial disease was HER2, if they recur a couple of years later you've got to know if it's still HER2 positive disease or not. So the question is since I would not use it in the future, would I use it in the current setting. And that's a very complicated question.

Obviously the easy answer is that there's not data to support the use of that, and at this point I would probably give you a call and see what you would do.

Dr. West: Thanks for the compliment, even if it was primarily a good hedge...

I'll tell you we were also wrestling with the issue that she had mediastinal involvement in multiple nodes that were completely invisible on the PET scan, which made us more concerned that she may have other occult disease that we are just now picking up. So the negative PET was not as reassuring as it would be if it correlated better with the pathology...

Dr. Hanna: Yeah, and the more I think about it, the more I would consider giving her some adjuvant erlotinib. If you consider that Erlotinib is the best drug we have against EGFR mutation positive disease, it is superior to chemotherapy. We know this. Logically it would make sense to give her the best drug for that biology.

So now that I've had a little bit of time to think about it, I probably would offer this patient erlotinib in an adjuvant setting.

I would not want to give it pre operatively though because we simply don't have enough information to know what the post operative risk would be in somebody who received erlotinib pre operatively. Could it increase the risk of ARDS [Acute Respiratory Distress Syndrome] post operatively? Could it cause pneumonitis? We just don't know that, so I would not give it pre-operatively.

Dr. West: I presume that that's entirely dependent on her actually having been tested for the EGRF mutation and being positive... because we also know from the IPASS trial that even though we reflexively think that an Asian never smoking woman with an adenocarcinoma means EGFR mutation, that's not the case.

Dr. Hanna: In Japan and in Asia it's not the case 40 percent of the time. In the US is probably not the case 50 to 60 percent of the time: we've learned our lesson, we have to treat based on biology, not clinical features.

(click on image to enlarge)

Dr. West: On a related note, I recently saw a patient who is about 60, another Asian never smoking gentleman who had adeno/BAC...

Dr. Hanna: You know Jack, in Indiana far more people have smoking histories: I'm not familiar with your patient population, but go on...

Dr. West: Yes, I sometimes have three never-smokers with lung cancer in a row in my clinic But like my question, is if you had a never-smoker who actually who had been started on erlotinib [Tarceva] before seeing you for a second opinion; they'd been on it for a week or two, and they hadn't been tested for the mutation, would you suggest a change or would you say let's watch and intervene if we see progression?

Dr. Hanna: The IPASS data would suggest if the EGFR mutation status is unknown, you should treat them with chemotherapy up front. If you look at the progression free survival [PFS] and the overall survival [OS] curves of the EGFR wild type patients who received gefitinib [Iressa], they progressed rapidly and some died very early. So you harmed them by not giving them chemotherapy up front.

If you look at the EGFR mutation positive patients, while gefitinib was better in terms of progression-free survival and appeared better in overall survival, while they did split, early on that wasn't the case. For the first 4-6 months the PFS curves and the OS curves overlap between the gefitinib and the chemo arm, suggesting that you're not necessarily harming patients early on by giving them chemo if they're mutation positive.

So the consequence of giving a patient who has EGFR wild type disease, gefitinib or erlotinib first line I think can be significant. But in a patient already started on it, you've already chartered a course. So what I would do is I would get a CT after 3 or 4 weeks of it. I'd want to know really early on if we're failing and switch promptly. I would not wait 2 or 3 months.

And frankly if the patient's responding, I would keep them on it, and more than likely they do have a mutation after all because the response rate with erlotinib in a wild type patient is about 2 or 3 percent. So most likely they would have a mutation.

Dr. West: It's interesting. I think about the IPASS data in the same context that we would think about the maintenance therapy data, in that in the maintenance setting, it's not that clear whether giving treatment earlier versus later is as important as just getting access to the right drugs over time. And I would say that the IPASS trial showed that progression-free survival showed a big difference depending on EGFR mutation status, but that the difference was blunted looking at overall survival...

Dr. Hanna: That's right...

Dr. West: Because you could cross over to the alternative. And if everyone ended up getting their chance with the right drug for them, people tend to come out very similarly.

Dr. Hanna: That's right, and treating a PS2 or PS3 patient with erlotinib is not hard. Treating a PS2 or PS3 patient with chemotherapy is really hard. So if you err and give them chemotherapy up front, you can probably make up for it by giving erlotinib second line.

[N.B. -- I believe that, like me, Dr. Hanna would be far less inclined to recommend adjuvant Tarceva now, after the negative results from the BR.19 trial that strongly trended toward a harmful effect of adjuvant Iressa, even in patients with an EGFR mutation.]