|
Home:
Web Guide 2: Program
Suplement: Richard M Goldberg, MD
DR RICHARD GOLDBERG SUPPLEMENT
DR NEIL LOVE: Welcome
to Colorectal Cancer Update. This is medical oncologist Dr Neil
Love. Every now and then a Phase III randomized clinical trial reports
results that have a major impact on standard of care. In breast
cancer, the research buzz this year has focused on the ATAC adjuvant
trial presented at the last San Antonio Breast Cancer Symposium
and recently published in the Lancet. In colorectal cancer, clearly
the most talked about study has been the Intergroup trial in metastatic
disease presented by Richard Goldberg at the recent ASCO meeting
in Orlando. This groundbreaking study reported a response rate,
time to progression and survival advantage for the so-called FOLFOX
regimen of infusional 5-FU/leucovorin and oxaliplatin compared to
two other arms – the Saltz or IFL regimen of CPT-11/5-FU/leucovorin
and a third arm combining oxaliplatin and CPT-11. Oxaliplatin has
just been approved by the FDA for use in combination with infusional
5-FU and leucovorin in patients who have progressed after receiving
irinotecan and bolus 5-FU. In fact, the approval time of seven weeks
is the shortest ever for a cancer drug in this country. Not long
after the ASCO meeting and before the FDA approval, I met with Dr
Goldberg to discuss the Intergroup trial data set, and he began
by commenting on the early evolution of the study’s design
DR GOLDBERG: 9741’s been
an interesting experience because it’s been a trial that’s
changed a lot during the course of the protocol. You can tell by
the “97” in 9741, that the concept actually began then.
The first iteration of the trial was a four-arm study looking at
5-FU/leucovorin as the control arm, and then three different ways
of giving CPT-11 with 5-FU and leucovorin. That actually started
in November of 1998.
By March of 1999, the NCI and we at Mayo were interested in looking
at combinations of oxaliplatin in colon cancer because of the promising
activity reports that were coming from Europe on this drug. The
way that we thought would be most expedient to test that in the
cooperative groups was to morph 9741 from a four-arm trial to a
six-arm trial. What we did was we actually added an oxaliplatin
plus CPT-11 arm and an oxaliplatin plus 5-FU/leucovorin arm, and
we subtracted one of the CPT-11/5-FU/leucovorin arms. So by March
of 1999, all of a sudden this four-arm trial became a six-arm trial.
What happened the following year was that we subtracted several
of the arms because of toxicity.
DR. LOVE: Can you talk a little
bit about what happened then?
DR GOLDBERG: We had instituted what we called
a rapid reporting mechanism to allow us to keep track of severe
side effects on the study. One of the problems is that the NCI toxicity
reporting forms only require notifying NCI of unexpected side effects.
What we found was that we were seeing patients who had a combination
of diarrhea and neutropenia and became septic, or had some sort
of thrombotic events that didn’t require NCI-rapid-reporting
forms to be filled out. So we implemented a one-page, faxable system
where people could notify us of any grade IV or V toxicity, or any
hospitalization that occurred on the study.
As a consequence of that, we were seeing death rates that were
really too high to be permitted, particularly in the arms were bolus
5-FU was being given either with CPT-11 or oxaliplatin. We dropped
those arms out of the study, so that we ended up collapsing the
six-arm study into a three-arm study. The three-arm study was what
we actually presented at ASCO this year.
DR. LOVE: Can you talk a little
bit about the toxicity that was seen? Was it something that was
unexpected?
DR GOLDBERG: Well, it was both expected and unexpected.
What we actually reported at ASCO in 2001 was a 4.5-percent 60-day
mortality rate for patients entered on the CPT-11/5-FU/leucovorin
or IFL regimen. What we were seeing in the two other regimens, the
oxaliplatin plus CPT-11 arm and the FOLFOX arm, the 5-FU/leucovorin/oxaliplatin,
was a 1.8-percent 60-day mortality. And Mace Rothenberg and a group
of people empanelled by the National Cancer Institute looked at
all of the charts on the patients that died within 60 days, and
several things became clear from that.
One was that about half of the patients died of something we expected
— dehydration, diarrhea, and neutropenia. But the other half
had thrombotic events, and these could include myocardial infarction,
cerebral vascular accident or even such things as mesenteric infarction.
This really identified a new syndrome that we hadn’t been
aware of before for toxicity of treatment for advanced colon cancer
patients.
The other thing that was interesting about this was, not surprisingly,
when you look at 60-day all-cause mortality compared to what investigators
report as treatment-related deaths, you find a difference. I don’t
think people are trying to hide things, but I think that all of
us are looking for reasons to think that it was disease and not
our treatment that led to a patient dying. This new metric that
we borrowed, actually, from the AIDS clinical trials’ literature
of 60-day, all-cause mortality, I think does provide a reliable
indicator of early toxicity that subtracts the impressions of the
treating physician out of the identification circumstances.
DR. LOVE: What do you think
the clinical implications are of this early toxicity that you saw
in terms of treatment with these regimens in a nonprotocol setting?
DR GOLDBERG: Well, I think, obviously, people
need to be really attentive to how their patients are doing. What
this did for us was to change our standard operating procedure not
only for patients being enrolled on this trial, but also for people
that we were treating off study. Instead of what I used to do, which
was to write four weekly treatments for people on IFL and see them
at six weeks, in the first course of therapy, I would see them weekly.
If I saw that they were having escalating problems with diarrhea,
nausea, vomiting, dehydration, I would reduce their doses within
the first cycle of therapy rather than sort of putting them on autopilot.
Also, I think, we were quicker to jump on problems and hydrate people
or hold treatment as a consequence of these findings.
DR. LOVE: Any specific things
or triggers that you look for, qualitatively, for you to stop therapy?
DR GOLDBERG: Well, I was looking for lack of resolution
of diarrhea when they came in for their next weekly treatment, and
I was looking for any signs that people were dehydrated at the time
of another therapy. Also, if their white counts were borderline
or low, that was another reason to consider either a dose delay
or a dose reduction.
DR. LOVE: Let’s go on
now and talk about the most recent results of this trial that you
just reported at ASCO.
DR GOLDBERG: What we did was reported on results
of approximately 800 patients who were entered on the trial. These
were patients who were simultaneously randomized either to the IFL
regimen, which was the new control arm of the study, to the FOLFOX-4
regimen, which is 5-FU/leucovorin/oxaliplatin, or to a regimen of
CPT-11 plus oxaliplatin. So, there were approximately 295 patients
per arm.
The North Central Cancer Treatment Group Data Monitoring Committee
actually released the results of this trial because the time to
progression and survival statistics surpassed the early stopping
rules that were written into the protocol. What I mean by that is,
there are rules called O’Brien-Fleming rules that have been
devised to try and tip data monitoring committees off to results
that are either strikingly positive or strikingly negative while
a trial is ongoing. Obviously, you don’t want to interrupt
a trial unless the results are striking. We had actually already
accrued all the patients to this trial that were needed, and so
what our Data Monitoring Committee actually did was to release results
early, rather than interrupt accrual.
Now, in retrospect, our data exactly recapitulated data from the
Saltz trial, which showed a 14-month median survival on IFL, and
also the data that Aimery de Gramont had presented on FOLFOX-4,
which showed essentially an 18-month median survival in his trial
as well. So, the data was consistent, it was reassuring, and it
was exciting.
DR. LOVE: How many patients
do you, yourself, have on the trial?
DR GOLDBERG: I would guess that I probably treated
about 15 patients on this study.
DR. LOVE: Did you get any hints
based on those patients or others that you treated of this kind
of activity?
DR GOLDBERG: I would say, yes, I did. We’ve
made a big commitment at NCCTG to the use of oxaliplatin, and so
I’ve had an opportunity to run a couple of Phase I trials
with oxaliplatin, CPT-11 and 5-FU/leucovorin, as well as to test
it in patients with pancreatic cancer. We’ve got a trial about
to start in esophageal cancer. We’d also done it in patients
with liver-limited colon cancer. The results that I was seeing were
striking. Responses were not only brisk, but also were to the point
where there was such rapid and dramatic shrinkage of disease that
it was like nothing I’d ever observed before.
DR. LOVE: It’s interesting,
I think we’ve become, with good reason, so sensitized to just
looking at randomized trial data and not just looking at individual
cases or anecdotal reports. But sometimes we do get hints of what
we’re going to see in individual patients. Any patient that
comes to mind that was either on the trial or not on the trial,
who demonstrates what you’re talking about?
DR GOLDBERG: Actually, there are a number that
come to mind, but one in particular is a 36-year-old speech pathologist
who had been found to have metastatic disease at the time of diagnosis
and had been seen at another center. There, she’d been given
a pretty bleak prognosis, told that she had about a year to live.
She ended up coming to Mayo, enrolling on this study, and being
enrolled on a FOLFOX regimen.
DR. LOVE: What was the site
of her disease and what was her condition when she was put on the
trial?
DR GOLDBERG: She had extensive metastatic disease
to the liver, to the point where she was evaluated by a hepatic
surgeon and was told that she wasn’t amenable to resection.
And this is an aggressive and very competent hepatic surgeon with
excellent judgment. She had had her primary resected at time of
presentation elsewhere. After about six cycles of the FOLFOX, she
had such a dramatic reduction in the number and size of her liver
lesions, that this same hepatic surgeon took her to the operating
room and, at the end of the surgery, she was disease-free.
DR. LOVE: How did she tolerate
the therapy?
DR GOLDBERG: She tolerated the therapy very well.
She continued to work, teaching school, during the time that she
was getting treatment. And she really didn’t have any severe
toxicity. One of the nice things about the oxaliplatin regimens
is they seldom cause significant alopecia. And, in our study, they
had a lower incidence of severe diarrhea, nausea or vomiting and
neutropenia than was observed with the IFL regimens.
She was able to work up until the time of her surgery, and then
was back at work six weeks later. She’s just passing her second
anniversary since resection, has no evidence of disease, and, obviously,
is very grateful. She had a six-month-old at the time of her diagnosis
and so far she’s been able to watch that toddler grow.
DR. LOVE: She hasn’t
had any post-surgical systemic therapy?
DR GOLDBERG: She has not.
DR. LOVE: Wow! What was actually
seen in her liver at surgery? What did the tumor look like?
DR GOLDBERG: Well, she just had three lesions.
It was clear that they were much smaller and they had a lot of necrosis
compared to what we would have expected based on her on-study scans.
DR. LOVE: What was her condition
when she started treatment? Was she having symptoms from the tumor?
DR GOLDBERG: No. Like most of the patients who
went on the study, most of them were performance score 0 or 1. And
she was a performance score of 0.
DR. LOVE: And she had no neurotoxicity?
DR GOLDBERG: She really had minimal neurotoxicity.
She had a little bit of the immediate-type neurotoxicity. For people
who have not had experience with oxaliplatin, there really are two
kinds of neurotoxicity that you observe. One occurs within 24 hours
of administration of the drug and is cold exacerbated. For instance,
people will describe that if they go to the refrigerator and take
a cold beer out the afternoon after their treatment, they’ll
feel like the beer can is electrified. Now, that usually disappears
within a day or two of treatment. But that can start occurring within
a few administrations of oxaliplatin. That isn’t so much the
problematic type of neurotoxicity.
The neurotoxicity that’s more problematic is a long-lasting
type of neurotoxicity that occurs usually somewhere after about
800 milligrams per meter squared of the drug. Since patients are
getting around 100 milligrams per meter squared of the drug with
each cycle, it doesn’t happen unless a patient is responding
to treatment, or at least stable with treatment, and gets a prolonged
course of therapy. What happens there is actually damage to dorsal
root ganglion cells. People can have a more prolonged, more typical
neuropathy, where they have difficulty buttoning buttons or picking
up a coin off the table. Now, different from what’s observed
with the taxanes or the vinca alkaloids, however, in most cases
this neuropathy improves over time, although it may not disappear
completely.
DR. LOVE: Let’s backtrack
into the trial data, maybe going back to this patient in terms of
how she was typical or how she was atypical. Can you talk about
basically what was seen in the three arms of the study?
DR GOLDBERG: What we observed was a significant
advantage both in terms of time to progression, which was the primary
end point of the trial, response rate and survival for the FOLFOX
regimen over IFL. And to give you hard data, the response rates
that we observed were a little bit lower than had been reported
in other trials. But I think response rates are often dependent
on the exact protocol that you follow. We saw a 28- and 29-percent
response rate for the CPT-11/oxali and IFL regimens, respectively.
And we saw a 38-percent response rate for the FOLFOX regimen. That
was statistically significant with a P value of 0.03.
But the other thing that we observed was the time to progression
data for IFL was 6.9 months compared to FOLFOX, which was 8.8 months.
And for the oxali/CPT-11 was 6.7 months. So, oxali/CPT-11 and IFL
were essentially the same. The P value for the comparison between
FOLFOX and IFL was highly significant, 0.0009. Again, that data
was pretty consistent with what was observed in the Saltz trial
of IFL and in the de Gramont trials of FOLFOX, but this was the
first comparison in a single randomized trial between those two
regimens.
DR. LOVE: How about the survival?
DR GOLDBERG: The survival actually was the most
interesting data, because the median survival with IFL was 14.1
months, and with FOLFOX, 18.6 months. With the oxaliplatin/CPT-11
it was intermediate at 16.5 months. Again, the P value of 0.002
for the IFL-FOLFOX comparison. And if you look at that in terms
of how many patients were alive at a year with IFL it was 58 percent,
with FOLFOX it was 71 percent, and with oxaliplatin/CPT-11 it was
65 percent of patients. We don’t have robust data at two years,
because our median follow-up on living patients is only about 14
months.
DR. LOVE: What about the post-trial
therapy that these patients received? Do we have information on
that and could that have had some impact in terms of survival?
DR GOLDBERG: Yes. I think it could have, and that’s
been, I would say, a criticism or a concern of this study. Obviously,
oxaliplatin has only been available in the United States on protocol.
Therefore, access to oxaliplatin in second-line therapy was relatively
infrequent for patients, whereas access to CPT-11 was essentially
possible for anybody.
If we look at the overall statistics, between 59 and 67 percent
of patients went on to second-line therapies. Among the patients
who were randomized to IFL, only 17 percent received oxaliplatin
second-line. In contrast, after the FOLFOX, there were 52 percent
of the patients that were given CPT-11, and after oxaliplatin plus
CPT-11, 45 percent of patients received 5-FU.
Now, after my presentation at ASCO, there was an interesting presentation
by a European investigator, Axel Grothey. What he presented was
data comparing the Mayo Clinic regimen, 5-FU/leucovorin to FOLFOX.
Again, he showed a survival advantage, a response rate advantage,
and a time-to-progression advantage for FOLFOX over 5-FU and leucovorin.
But the other thing he did, which was quite interesting, was he
put up a slide which looked at the six most recent randomized trials
presented at ASCO in advanced colon cancer, and showed that when
more patients had access to all three drugs, the survival went up.
For instance, in the Saltz trial, only about five percent of patients
were able to get oxaliplatin second-line and the median survival
in that trial was just over 14 months.
In the trials from Europe where oxaliplatin and CPT-11 were available
for second-line therapy, the median survivals were running between
19 and 21 months. His point was similar to the point that I had
made in my discussion, which is that the optimal therapy for patients
with advanced colon cancer is to have all three drugs available
— 5-FU, CPT-11 and oxaliplatin. Exactly how to sequence them
is really still up in the air, but I think the sequencing is less
important than the availability of all three drugs.
DR. LOVE: What about toxicity
in the three arms?
DR GOLDBERG: Well, that’s an interesting
issue, as well, and one that I’ve already alluded to. Actually,
more patients went off FOLFOX for toxicity, than went off IFL or
oxaliplatin/CPT-11 for toxicity. However, this occurred usually
late in treatment, after about eight or so cycles, and most of the
patients who went off, went off for neurotoxicity while responding
to therapy.
The toxicity profile for the IFL was exactly the opposite. Patients
had early toxicity — most often diarrhea, nausea, vomiting
and dehydration — and required either dose reductions, were
taken off of treatment, or some died of toxicity, early on. So,
the difference is that the IFL toxicity is a little bit less predictable,
occurs earlier and, in my opinion, is more potentially life threatening
than the toxicity that was seen with the FOLFOX regimen.
DR. LOVE: Now, the time to
progression, when a patient went off therapy who was responding,
were they considered progressing?
DR GOLDBERG: That’s been another interesting
point of controversy. The definition that we used of time to progression,
essentially, is the same definition that we’ve used over 25
years in NCCTG. And, obviously, this was an NCI-supported and NCI-scrutinized
trial. So, our definition for time to progression had been reviewed
by the NCI. Patients were not considered to have progressed until
they had a CT scan that showed progression, or some other assessment
of their measurable disease. So it is possible that a patient could
have gone off the FOLFOX regimen for toxicity and been treated with
CPT-11, and some of the benefit from CPT-11 could have actually
been accrued to the FOLFOX regimen.
Now, we actually looked at how many patients that applied to. What
we know is about half of the patients who got FOLFOX went on to
get CPT-11, as I mentioned. About half of those patients got CPT-11
within three months of the time that they went off of therapy, and
about half of those patients had a response to therapy. So, really,
only about one out of every eight patients that went off of FOLFOX
for toxicity, got a benefit from CPT-11 early on, after they went
off FOLFOX.
We did actually re-analyze our study, looking at the same definition
that Pharmacia had used in their FDA presentation of time to progression.
And we found that there still was a significant time-to-progression
advantage for FOLFOX over IFL.
DR. LOVE: To what extent is
the cumulative toxicity of oxaliplatin reversible?
DR GOLDBERG: Well, that’s an interesting
question that I can’t really address from the data from our
study yet, but I can address it from data that Aimery de Gramont
presented at ASCO in a poster. He’s the chair of a trial that
he’s termed the MOSAIC Trial, and that’s an adjuvant
study in which patients were randomized to 5-FU/leucovorin or 5-FU/leucovorin
plus oxaliplatin. What he found in that study was that neuropathy
was quite frequent, particularly in patients around the ninth, tenth
and 12th cycles. The patients got 12 cycles of therapy over half
a year’s time, and many of those patients did get neuropathy.
But in most cases that neuropathy either disappeared or eased dramatically
within four months of going off treatment.
DR. LOVE: Is that your clinical
observation?
DR GOLDBERG: My clinical observation has been
that the neuropathy decreases over time, although I think that it
remains more problematic, in my experience than what Dr de Gramont
presented.
DR. LOVE: Now the patient that
you talked about before, the young woman, had minimal or really
no neurotoxicity. How often do you see that with oxaliplatin?
DR GOLDBERG: Well, it depends, again, on how many
treatments the patients get. About 50 percent of patients who get
out to 10 or 12 cycles of therapy will have some significant amount
of neuropathy, although seldom is it severe. Seldom is it grade
III or higher neuropathy. And many patients will say, “I want
to keep on treatment, because response is more important to me than
this symptom.”
DR. LOVE: If you take a step
back from the clinical data and start to talk about it from the
point of view of the clinician, in terms of quality of life for
these patients, how would you compare FOLFOX to IFL?
DR GOLDBERG: Actually, our trial does have a formal
quality-of-life analysis built into it, but as you might expect,
there’s a tremendous amount of data associated with that.
We haven’t been able to analyze that data yet. We’re
hoping to be able to analyze that data in time for an abstract submission
to next year’s ASCO. However, what I would say is, initially
at least, the toxicity with the FOLFOX regimen, I think, is less
severe than it is with the IFL regimen. And I think it’s particularly
noticeable with respect to the hair loss and diarrhea, as well as
nausea and vomiting. So, in my opinion, the statistically significant
advantage that we saw in terms of toxicity for FOLFOX was also a
clinically relevant difference.
DR. LOVE: At this point, of
course, oxaliplatin is not available. Hopefully, it’s going
to be available, I guess, hopefully by the end of this year. If
it were available, how would you see it being integrated into nonprotocol
therapy?
DR GOLDBERG: That’s an interesting question.
I was actually in Europe last week, speaking with investigators
from the EORTC, and they’ve obviously had access to all three
drugs for quite a lot longer than we have. And I polled the audience,
who was mainly made up of oncologists from Europe, about who uses
CPT-11 regimens first, and who uses oxaliplatin regimens. And actually,
I would say it was divided about 50-50.
Now, I have a lot of respect for Aimery de Gramont, who has essentially
coined both the FOLFOX and FOLFIRI regimens. I had an opportunity
to discuss with him what his current strategy is, and he basically
uses oxaliplatin as his first-line therapy in virtually everyone,
unless they have symptomatic neuropathy to start with. What he’s
taken to doing is giving a limited number of cycles of FOLFOX, around
six or eight, and then immediately switching over to FOLFIRI even
though patients may be responding or stable. Then, when they progress
on the FOLFIRI, he’ll go back to FOLFOX. With that, he thinks
that he can optimize the neuropathy. Since the neuropathy will often
disappear, if you have a holiday from the oxaliplatin, you can get
in a higher amount of that drug over the entire length of the patient’s
treatment for advanced disease. He’s projecting that, with
that strategy, he hopes to see median survivals exceeding two years
with the drugs that we have available today. That, to me, is dramatic
news, considering that with no treatment median survival for advanced
disease is around six months and, with 5-FU and leucovorin, it’s
only around 12 months.
DR. LOVE: What about other
combinations with oxaliplatin? One I’ve heard talked a lot
about is capecitabine and oxaliplatin.
DR GOLDBERG: Well, I think that that’s a
very interesting potential regimen. There was some data presented
on Phase II studies in Europe that show promising activity of capecitabine
plus oxaliplatin. I think, in the U.S. at least, there’s some
reluctance to embrace infusional 5-FU regimens. The reluctance is
based on the need for an indwelling catheter, because of catheter-related
complications, and also, quite frankly, because catheters and pumps
malfunction and cause patients, as well as their managing physicians,
problems in the middle of the night. If we could replace the infusional
5-FU with an oral drug that mimics infusion, it would be ideal.
DR. LOVE: Any guesses as to
whether that’s going to be possible?
DR GOLDBERG: Well, I think the randomized trials
need to be done, looking at that, but I’m hopeful that it
will be possible.
DR. LOVE: In the interim while
oxaliplatin’s not available in the United States, what’s
been set up in terms of clinicians obtaining it?
DR GOLDBERG: Well, unfortunately, the supply of
oxaliplatin that’s available in the United States is currently
limited. The reason for that is, even though there’s plenty
of oxaliplatin being produced in Europe, the factory that makes
it has not been inspected by the FDA. Since that factory is fully
booked for drug development, they aren’t really interested
in inviting the FDA in. The factory that manufactures the drug in
the United States has to obtain its raw material from a factory
in Japan, and there’s a limited amount of that raw material
that can be imported into the United States as long as this drug
is an investigational drug. Once it becomes widely available and
is no longer an investigational drug — meaning it’s
FDA-approved — there should be an unlimited supply available.
Until that time, the NCI has developed a lottery, similar to the
lotteries that were used for taxanes early in their course. Patients
have to be referred to a cancer center and then will be able to
get entry into a lottery. The lottery is restricted to patients
with first-line therapy for metastatic disease. So, it’s really
quite limited. Now, as has been the case over the years, patients
could obtain the drug in France or Europe or Mexico, and get an
oncologist here to administer it to them, if they have the resources.
DR. LOVE: What are some of
the more, let’s say, interesting or provocative comments or
reactions you’ve had since the ASCO meeting or during the
ASCO meeting, to this trial? DR GOLDBERG: I think there have been some skeptics
about the findings of this trial, and many people have asked me
if I think the reason we’re seeing this difference in survival
is because IFL uses bolus 5-FU and the FOLFOX uses infusion 5-FU.
My response to that is, there have been many trials comparing bolus
5-FU to infusion 5-FU. In no single trial has there ever been a
statistically significant survival advantage to infusion over bolus,
although there’s often been a trend. There was a meta-analysis
published in the Journal of Clinical Oncology of over 2,000 patients
randomized to bolus versus infusion, and that only showed a modest
advantage of about a month to the infusion over the bolus regimen.
When we’re seeing a 4.6-month difference in median survival
between FOLFOX and IFL, I don’t think that you can possibly
ascribe all of that difference to the difference between bolus and
infusion 5-FU.
|
