The Biology of Extending Survival in Advanced CRC - Episode 8
John L. Marshall, MD: The plenary session paper at the 2017 ASCO Annual Meeting on colorectal cancer is an adjuvant paper—no new drugs. It is a treatment pathway. Dirk, why don’t you run us through what the analysis is and what we’re seeing.
Dirk Arnold, MD, PhD: It’s a combined analysis of 6 trials, which have been run over several years in several healthcare systems at several scientific groups—Europe, the United States, across the world. And the question was very simple. It was: should FOLFOX, as adjuvant treatment, or capecitabine/oxaliplatin be given for 3 months or for 6 months? And if it would be given only for 3 months, what kind of loss of efficacy would be seen? Is it worth calling this noninferior or calling this similarly efficacious?
John L. Marshall, MD: I was looking for 2 endpoints. I was looking to see, are the curves overlapping? What’s the neurotoxicity? Do I reduce toxicity? Does anybody want to take a stab at what we found?
Fortunato Ciardiello, MD, PhD: Clearly, I think it is obvious that if you do less therapy, you will have less toxicity.
John L. Marshall, MD: So, that part of my hope was positive, right? I think it’s about a 50%, 60% grade 3/4 neurotoxicity versus less than 20%.
Fortunato Ciardiello, MD, PhD: You have to consider that neurotoxicity is really a major problem for these patients. We always say it disappears quickly when you stop it, but this is not true in reality. A lot of patients still experience some neuropathy for a long time after they finish it. So, if we can have the same efficacy as adjuvant therapy that we do to all patients, we don’t know if they really need that or not, because this is the problem when you do adjuvant therapy, even in stage 3 disease.
John L. Marshall, MD: So, the toxicity was better?
Fortunato Ciardiello, MD, PhD: The toxicity was better, and I think this is good news. Now, the question is, is that as effective as more toxic? Because, in oncology, we are always saying, “The more toxicity, the more activity.” That translates to more efficacy. So, when you have a drug that is less toxic, you are afraid. And now we have a treatment that is shorter. Is this shorter treatment like a priming?
John L. Marshall, MD: We’re talking about curative therapy, too. Every patient matters. Each one of us will see 50 to 100 adjuvant patients a year, and if there’s a 1% difference, 1 of our patients will die if this is wrong, right?
Fortunato Ciardiello, MD, PhD: Yes.
John L. Marshall, MD: Even in the 1%, there’s the emotional attachment. We know in the sub trials that piled together, one was basically negative. It said that 6 months was a little bit better. Another one was a tie, the pooled analysis.
Dirk Arnold, MD, PhD: The pooled analysis was showing that noninferiority could not be confirmed. Within the assumptions that were made, we’re always compromised between clinical reality and clinical wish and what statistics can prove.
John L. Marshall, MD: How tight those curves get.
Dirk Arnold, MD, PhD: How tight can it really be? But noninferiority, formally, was not a problem. Therefore, we can formalistically say, “Yes, we have to continue 6 months because the noninferiority within the defined experiment was not seen.”
John L. Marshall, MD: But you’ve seen the curves.
Dirk Arnold, MD, PhD: However, the curves look, let’s say, superimposable.
John L. Marshall, MD: You couldn’t put a piece of paper in between those curves.
Dirk Arnold, MD, PhD: Yes, but for some of the subgroups, you could, but generally.
Paul R. Helft, MD: That’s one of the important questions that was raised in the more locally advanced patients who have deeper T stage tumors, or more nodes that are positive.
John L. Marshall, MD: T4 N2.
Paul R. Helft, MD: Exactly, right. And the curves tend to separate in that subgroup analysis, again, with a large number of patients being examined in that subgroup. I think we still will be left with a few questions about that.
Tanios Bekaii-Saab, MD: It’s disease-free survival.
John L. Marshall, MD: Yes, it’s a 3-year disease-free survival, which is a surrogate marker.
Tanios Bekaii-Saab, MD: But the more advanced patients are the ones that have the likelihood of being, already, metastatic. You may be delaying the disease reoccurrence, but not necessarily.
Paul R. Helft, MD: Without an absolute risk reduction.
Tanios Bekaii-Saab, MD: Not changing the risk of death. It’s interesting. One of my biggest challenges when we were part of CALGB 80702 was that whenever I presented patients with a 3-month versus 6-month option, they chose 3 and they didn’t want to go on study.
John L. Marshall, MD: Yes, the “3” were nervous patients.
Tanios Bekaii-Saab, MD: Even though the patients understood that there may be a slight loss of benefit with the 3-month regimen, they still opted for the 3. My compromise had been to go for 3 months of FOLFOX or CapeOx (capecitabine and oxaliplatin) straight and then the other 3 months were just simple 5-FU or capecitabine. We know that 5-FU is where you get most of the benefit, and not from oxaliplatin.
John L. Marshall, MD: There’s no question, that’s true.
Paul R. Helft, MD: There’s a psychological barrier that we have about less treatment as oncologists. I’m about 50 years old, and I don’t remember when we gave more than 6 months of therapy.
Paul R. Helft, MD: Of course there was a trial in the late 1980s that demonstrated that 6 was equivalent to 12, and I wondered how you felt, at the time, about it?
Fortunato Ciardiello, MD, PhD: This had to do with the adjuvant treatment of breast cancer. They showed that 12 months and 6 months were equal. I have, also, another psychological consideration. For a patient to stop after 3 months, it means that the story is over and the cancer is cured—or is not. The patient is just a cancer survivor. And 6 months means another 3 months of coming to do therapy. Because the patient thinks he or she has to be sick when on treatment, it becomes psychologically devastating for the follow-up. But he’s not sick. He goes back to life.
John L. Marshall, MD: Yes, but there is that security blanket of being on treatment. Patients are more comfortable.
Fortunato Ciardiello, MD, PhD: Yes, it’s very complex. But this is adjuvant therapy, no?
John L. Marshall, MD: Yes, adjuvant. The T4 N2s were pulled out and they did worse—maybe, there is some suggestion. But the other group, the T3s and the N1s, did very well and there was really no difference in that subgroup in terms of 3 versus 6. Maybe we’re starting to decipher if it will matter, a little bit, there? And I’d be willing to show a patient those curves to reassure them that 3 was just as good—“Look at these curves.” Does anybody have thoughts on that?
Fortunato Ciardiello, MD, PhD: As a consideration, John, every time in a crisis like this one, we have some statistical considerations. The facts, at some interpretation, are very complex. You don’t know where to draw the line to be on one side or another. This opens up, again, our role as doctors and as physicians. And the communication with individual patients becomes the center of our life. We will be there, in front of our patients, to say, “These are the facts. You’re being treated and, maybe, you’re being cured. We know that you need some adjuvant therapy. We can offer you this.”
We can balance together and decide, for example, on a compromise. It is common sense. For me, after these data I would be making a decision about one of the options I will do for patients, but then discussing with the patients—again, without delegation—everything else. I think this is a take-home message for us. We should be physicians interacting with patients.
Tanios Bekaii-Saab, MD: We’re not statisticians, we’re physicians.
John L. Marshall, MD: Dirk?
Dirk Arnold, MD, PhD: That’s the amazing thing. There is a highly-sophisticated debate about clinical use about statistics—about assumptions. But if I look to my daily practice, at what I have been doing during the last 4 years, there is no change and there is no gain in insight. What we have been doing is, when patients do develop neuropathy after 3 to 4 months, we just discontinue oxaliplatin and let them potentially run with 1 or 2 remaining months with 5-FU, alone, to prevent this neuropathy. This is reinforced.
John L. Marshall, MD: So, that practice is reassuring?
Dirk Arnold, MD, PhD: Absolutely.
John L. Marshall, MD: You can tell a patient that you can stop oxaliplatin. It does drill down on the question of, should you continue the 5-FU? This is cost. This is all of the things that go with it. I have several patients who know that this data is being presented, and I’m going to face them next week in clinic. If the Washington Post doesn’t cover the story, then I’ve got to, for them. What should I tell my 3-month T3 N1 patient right now? Are we done? Can I tell them we’re done?
Tanios Bekaii-Saab, MD: Yes.
Paul R. Helft, MD: Yes.
John L. Marshall, MD: Yes?
Fortunato Ciardiello, MD, PhD: Yes.
John L. Marshall, MD: Now, you’re going to keep the Xeloda (capecitabine) going, aren’t you?
Tanios Bekaii-Saab, MD: Again, I think it’s reasonable. As Fortunato said, it’s very reasonable to have that discussion. But I think you have to have the 2 discussions. You have to have the discussion of, “Listen up, it’s very likely that you don’t need any further chemotherapy” with the T3 N1 patient. But, if you feel the anxiety of the patient—essentially, if the patient feels that maybe another 3 months with less intense therapy is better—I think it’s reasonable, then, to talk about 5-FU or capecitabine. And for the T4 N2 patient, I’d have the same conversation. I think the data is compelling enough, in my mind, that 3 months of FOLFOX or CapeOx is more than enough for all patients. The assurance is the extra 3 months of fluoropyrimidine. That is the bulk of the discussion. It’s not the first 3 months, it’s the next 3 months.
Transcript Edited for Clarity