|
Home:
Web Guide 2: Program
Suplement: David J Kerr, MD, FRCP
DR DAVID KERR SUPPLEMENT
DR NEIL LOVE: Dr
Venook referred to European studies looking at combinations of capecitabine
and CPT-11 and another ASCO presentation by Dr David Kerr focused
on a Phase II trial looking at this regimen. I met with Dr Kerr
to learn more about this new data set and he described the rational
for combining these two agents.
DR DAVID KERR: Both drugs are active as single
agents in their own right in colorectal cancer, and there’s
a lot of information from preclinical models, in vitro models, suggesting
that they may be truly synergistic when brought together, which
is something that, as clinicians, we’re always, always looking
for. And it does make compelling sense to bring these two mechanistically
novel drugs together into a single combination. Working with colleagues
in the Netherlands, we’ve entered about 27 patients, going
from Phase I into Phase II. These are chemotherapy-naïve patients
with advanced measurable colorectal cancer. By doing a fairly conventional
dose-finding Phase I study, we’ve come up with a recommended
dose of Xeloda at 1,000 milligrams per meter squared BD daily, given
by mouth for 14 days, and irinotecan given as a brief 30-minute
IV infusion at a dose of 250 milligrams per meter squared given
every three weeks. So, the combination’s a three-week repeat
with the infusion of Camptosar on day one, starting the Xeloda on
day one also, but for 14 days, a seven-day break, and a three-week
repeat.
At this recommended dose level, we have a response rate of around
about 50 percent, which is in the ballpark that we would expect
with this sort of combination. We have tolerable side effects. Dose-limiting
toxicities are myelosuppression and diarrhea. But at the recommended
dose, we would have projected grade III-IV toxicity rate of around
10 percent. So, we think this is manageable toxicity. It’s
active, and it’s very convenient for patients.
DR. LOVE: Can you talk a little
bit more about biologically and pharmacologically exactly why you
would expect there to be synergism?
DR KERR: When we put combinations of drugs together,
it’s always attractive to use drugs that have a different
molecular target. For Camptosar, or irinotecan it’s topoisomerase-1,
and clearly for Xeloda, which is a prodrug 5-FU that feeds in through
inhibition of thymidylate synthase. And one would imagine that the
drug, Xeloda, which starves the cell of thymidine, the building
blocks of DNA, might synergize with the drug, irinotecan, which
inhibits DNA topoisomerase-1 to be effective if the cell is depleted
of thymidine, then there’s no doubt that one would expect
that the camptothecan, irinotecan, would be much more effective
in that setting.
DR. LOVE: Any reason to think
that Xeloda would be a better combination with irinotecan than 5-FU
or 5-FU/leucovorin?
DR KERR: I think when we compare Xeloda single-agent
with some of the 5-FU/leucovorin combinations out there, I have
a distinct feeling that Xeloda is much more useful than the frequent
bolus administrations of 5-FU/leucovorin. I think that’s been
shown in some very compelling randomized studies in advanced disease.
There are toxicity savings, convenience savings, and patient preference
— all that those things are loaded toward Xeloda.
What is interesting, in Europe, we much more commonly use infusional
5-FU and leucovorin compared to common practice in the States. And
there are no direct competitors in that setting, and that would
be an interesting one. But my guess is that Xeloda would be probably
as efficacious and probably with about the same level of toxicity
as some of our rather more complicated 48-hour infusional 5-FU/leucovorin
regimens. The problem with the complicated regimens is — the
infusional ones — you require pumps, you require intravenous
access. And it’s difficult for the patients and it has its
own associated morbidity. I think that if we can substitute Xeloda
in these combinations with drugs like CPT-11 or oxaliplatin, then
that does seem to be the way ahead.
DR. LOVE: I’ve heard
people say that Xeloda is essentially similar to or equivalent to
infusional 5-FU, and I’ve heard other people say, no, that’s
much too simplified. What’s your thought on that?
DR KERR: I think that it’s a useful rule
of thumb, but like any statement, until we can back it up with the
evidence of a randomized trial, it will always sit between us as
something that seems reasonable, intuitive, but there’s no
hard evidence backing it up. And there are some moves in Europe,
for example, to do some comparative studies using our combination
of Xeloda and irinotecan versus infusional 5-FU/leucovorin and irinotecan.
And so the proof of the pudding will be in that eating.
DR. LOVE: Well, I guess I was
thinking, obviously, we don’t really have the answer clinically
in terms of a randomized study, but pharmacologically do you think
that essentially it’s the same thing? Because it really isn’t
— you’re 14 days on; you’re seven days off. It’s
not quite the same.
DR KERR: No. I think that in pharmacologic terms,
in terms of the degree of inhibition of thymidylate synthase and
so on, I think there are some data suggesting that they are approximately
equivalent. So, I think, on those grounds, from small, detailed
patient studies looking at pharmacodynamic end points, they do seem
similar.
DR. LOVE: Hmm. How does this
study compare to other studies that have been done of the same combination?
DR KERR: Interestingly, there are two or three
other European studies which have come to very similar conclusions
in terms of dose-finding. And that’s great, because it always
adds certainty to any clinical observations that we make. There’s
an important French study by Delord, there’s another German
study that has come to very similar conclusions, and an Italian
randomized Phase II study. So, the dose of irinotecan at 250 per
meter squared and Xeloda with a total dose of 2,000 milligrams per
meter squared daily for 14 days does seem as if it’s the one
that’s going to move forward into all the big Phase III randomized
studies.
It’s important to get it right now, because clearly, if we
go with a single Phase I and Phase II and establish all of the Phase
III programs on one Phase I study, then if you get it wrong, that
can come back to haunt you with excess toxicity. And there are some
resonances with some of the recent experience with the bolus 5-FU/leucovorin/irinotecan
combinations, which have been used in the States. So, the fact that
we have got really four studies now, showing very similar conclusions,
adds weight and certainty to that regimen.
DR. LOVE: I wonder if you could
think through or pull out in your mind a case of a patient who was
in the study that you’re presenting here and present it to
us as an example of how you think through the decision of management
of metastatic colorectal cancer and how this trial played out in
that patient?
DR KERR: One interesting person that we could
present was somebody who participated in the Phase I trial with
the capecitabine and irinotecan. This is a patient who had hepatic
metastatic colorectal cancer, and it was deemed inoperable at that
stage. He had bilobular disease, and 25 percent of the liver was
replaced by tumor. After a multidisciplinary team meeting, it was
very clear that he had inoperable disease.
DR. LOVE: How old was he?
DR KERR: He was an ex-Army brigadier in his late
sixties.
DR. LOVE: And was this at initial
presentation or he had been treated earlier?
DR KERR: He had presented some two years before
with Dukes’ C colorectal cancer. At that stage, he hadn’t
received adjuvant chemotherapy, strangely. He was treated in a different
center, and he re-presented to our center with evidence of advanced
hepatic metastatic disease.
DR. LOVE: Did he have disease
anywhere else?
DR KERR: No. Lungs were clear, as far as we could
tell from CT and MRI scanning. The rest of his abdomen was clear,
too.
DR. LOVE: What was his condition
at that point?
DR KERR: His performance status was 1, but he’d
always been a fit and active man, so he brought that soldierly strength
and bearing with him. So, he was in quite good shape, but again,
given the way that soldiers are, I guess hiding a lot of the symptoms.
He was entered at dose level 3, which is a combination at that stage
of irinotecan 250 milligrams per meter squared and Xeloda 2000 milligrams
per meter squared daily through days 1 to 14. And we repeated CT
scans after two months, and he had an excellent partial response
by that stage. We gave treatment for another two months repeated
the CT scan and, at that stage, it was obvious that his disease
was now unilobular. He only had two deposits left out of what were
10 initially. We decided that we would go ahead and resect the disease.
So, there’s an example of using chemotherapy to downsize a
tumor that was initially inoperable, to the stage of operability.
DR. LOVE: How did he tolerate
the therapy?
DR KERR: He tolerated the treatment well. He
didn’t have to undergo any dose reductions at all throughout
that four-month period. He had noticeable thinning of his hair,
but there were no treatment delays, nor were there any dose reductions
at that dose level.
DR. LOVE: How was his performance
status and quality of life affected by therapy?
DR KERR: His performance status was, I think,
well maintained. He was an active man. He was chairman of a local
hospital trust; he kept that going and lived in a large house with
a garden that he kept going too. In a sense coming up to the hospital
once every three weeks, rather than, certainly in Europe, with the
rather more frequent complicated regimens that we use with infusions,
did free him up. And there’s no doubt that he appreciated
that.
DR. LOVE: So where is he right
now? What’s happening?
DR KERR: He underwent resection of residual disease
two months ago. He came through the operation very well. His liver
has started to regenerate and his most recent CT scan two months
postoperatively was completely clear. All tumor markers are normal.
DR. LOVE: Did they biopsy or
look at other parts of the liver? I’m sure they looked.
DR KERR: The surgeons performed an intraoperative
ultrasound and found nothing at all, so there was nothing for them
to biopsy. All tumor markers were normal. And we would just keep
him under a regimen of regular follow-up from now on.
DR. LOVE: Now, were his tumor
markers elevated before this?
DR KERR: Yes. His CEA level was 10 times the
upper limits of normal. That came down very quickly within the first
two months of treatment, to normal levels, in keeping with the excellent
partial response that he had. So it’s a situation which, I
think, increasingly around those patients that come with isolated
hepatic metastases, with these powerful chemotherapies with high
response rates, we may be in a new ballgame of actually downsizing
rather than down-staging, patients with hepatic metastatic disease.
I think that we should operate on them, because we’re moving
into a very different natural history of the disease.
DR. LOVE: Are you planning
on restarting chemotherapy?
DR KERR: No. He had his four months of treatment
and he had that excellent response. He’s had residual disease
resected. As far as we can tell, there’s no residual tumor.
The likelihood is, of course, that there is, but we would give him
a period of time off treatment and just merely follow him up.
DR. LOVE: Are you thinking
that he’s cured?
DR KERR: No. I think he may be part of a group
of patients now who have a five-year survival rate of about 15 to
20 percent, which is far from cure. I believe that if we hadn’t
operated to remove residual disease, then his median life expectancy
from that point would have been 12 to 18 months.
DR. LOVE: So, you are assuming
that the tumor that was in the other parts of his liver is probably
still there?
DR KERR: I’m assuming that, but there is
a possibility that he may have had a complete pathologic response.
I say “possibility,” because we didn’t check it.
And, therefore, we are moving into a new period of the natural history
of management of this disease, because this isn’t something
we’ve done before. Of course, there are neoadjuvant data with
infusional 5-FU/leucovorin/oxaliplatin that look very interesting,
too. But from quite large surgical series collected from the hospital
Paul Brousse in Paris, in those patients who had initially inoperable
disease that was then made operable by giving adjuvant chemotherapy,
their follow-up suggests that the survival for those patients is
as good as those patients who had operable disease right from the
beginning. That’s not a randomized trial, but it’s a
sophisticated anecdote from one of the largest surgical series in
the world.
DR. LOVE: Do you think there
is a role for a randomized trial? Are there enough patients like
this?
DR KERR: Definitely. Definitely. Hepatic metastasis
is the common site of morbidity and mortality with colorectal cancer.
More than 85 percent of patients die with hepatic metastases. Ten
to 20 percent will present with disease microscopically confined
to the liver. I’ve just finished a trial in the U.K. with
300 patients, randomizing intravenous versus intrahepatic arterial
chemotherapy for patients with disease confined to the liver. So,
the patients are out there. They are being referred to our community
at an earlier stage, and definitely we should be doing trials of
neoadjuvant treatment looking at these new combination chemotherapies.
Definitely.
DR. LOVE: That was a very interesting
case. What’s the reaction been of this man and his family
to this experience?
DR KERR: Excellent, really. It’s rewarding
for us, too, because dealing with and using drug combinations with
high response, it’s fantastic. And I’ve been involved
in colorectal cancer treatment for 15 years, and when we kicked
off, it was absolutely miserable. Very few medical oncologists were
interested in it, and there were very few of us that were involved
in those early days of testing the new drugs and all the rest of
it. It's very different now. The colorectal cancer trials are amongst
the most rapidly recruiting in the world. There are huge numbers
of meetings associated with them. Much more interest in developing
ranges of novel biological therapies and so on, so it’s a
very exciting time to be an academic interested in colorectal cancer.
But it wasn’t always thus.
DR. LOVE: I know one of your
interests has been gene therapy for colorectal cancer. Can you talk
a little bit about that?
DR KERR: It’s an interesting time for gene
therapists after all the much heralded initial brouhaha aired on
gene therapy curing cancer and all the rest of that nonsense, of
course. There’s no doubt that the third to four generation
viruses that we’re using now, I think, have real clinical
potential and merit. In terms of our own receptor program in Oxford,
we’ve taken the common cold virus, the adenovirus, and manipulated
it so that the virus can divide, but only in cancer cells, those
which have got mutant P-53 or some other molecular switches. What
we do is, we can actually plumb into the virus. In terms of us moving
genes around, it’s almost as easy as copying and pasting word-processing
documents. So, we can put new genes in. These are genes —
these are enzymes that convert nontoxic, inactive prodrugs to very
cytotoxic species.
The one that we’re interested in is something called nitroreductase,
and it can convert a prodrug called CB1954 to a very toxic bifunctional
alkylating agent that is active in all the colorectal cancer cell
lines that we have tested it in. And we’ve just completed
our first clinical trial, not published yet, in which we’ve
injected the virus directly into the cancers of patients with hepatic
metastases that have come out through operation. What we can see
is very significant expression of the virus and of the enzyme.
So, that was a proof of principle. And were going on to combination
studies now with the virus, the prodrug, probably in combination
with chemotherapy, because we’ve shown that our virus-directed
enzyme prodrug therapy (VDEPT) synergizes with 5-FU/leucovorin and
with Xeloda. Therefore, one can imagine the possibility of perhaps
someday in an adjuvant setting, us giving our complicated postoperative
adjuvant chemotherapy, but perhaps also giving some adjunct to viral
therapy, gene therapy, to the liver. Perhaps it will be administered
through the portal vein or into the peritoneal cavity, other sites
of recurrence of the tumor, and adding value to the existing treatments
with these very molecularly-focused therapies that we can target
to the liver and to the peritoneal cavity. So, it holds some promise.
That’s ten years away, but that’s where I see us going.
DR. LOVE: Wow, that’s
a fascinating strategy. Has this concept of using gene therapy to
basically activate — if I understand what you’re saying
correctly — a cytotoxic agent been used in other tumors or
by other investigators?
DR KERR: Yes. There are three broad examples
of this. One is, we clone an enzyme from a bacterium or from a virus,
so there’s no human equivalent of it. And we then put that
— the gene from that enzyme — into a viral vector, adenovirus,
in our case. So, we’ve got NTR and CB1954. That’s the
bifunctional alkylating agent. We’ve also worked with cytosine
deaminase, another bacterial enzyme that can convert the antifungal
drug, 5-fluorocytosine to 5-FU, at very, very high intracellular
concentrations. And there’s another enzyme called thymidine
kinase that can convert ganciclovir, again, to a very toxic antimetabolite.
And so you can imagine, we’ve done this.
You could actually almost have combination chemotherapy from one
virus. We’ve made a virus that’s got two enzymes in
it, one for NTR, the alkylating agent, one for cytosine deaminase,
and we’ve shown that you can get synergistic cell kill by
generating both 5-FU and the alkylating agent within the same cell.
So, we can deliver combination chemotherapy from a single virus.
DR. LOVE: Are there other investigators
using similar strategies?
DR KERR: I think we’re sort of leading
in terms of the colorectal cancer stuff. The nitroreductase is something
that we cloned and developed with colleagues in the U.K. So, we
are the prime leaders in that. There’s some work going on
elsewhere with cytosine deaminase and 5-fluorocytosine, but in terms
of all this combination work, I think we’re the leading group
in that sense.
DR. LOVE: Now, you talked about
injecting it directly into a hepatic metastasis. Ultimately, are
you thinking that this is going to be administered systemically?
DR KERR: There are a number of stages in developing
the gene therapy. So, proof of principle was direct injection into
the metastasis. Does a virus infect the cells? Is the enzyme made?
Proof of principle. We’ve done that. There’s a tick
in that box. But giving the viruses through the hepatic artery or
portal vein in an adjuvant setting could be a clever thing for us
to do, or directly into the peritoneal cavity. I’m a great
supporter of regional chemotherapy. I’m interested in the
pharmacokinetics and pharmacodynamics of it. There seems logic to
doing that. But there’s no doubt that we, and many others,
are working on trying to develop stealth viruses that we can give
intravenously because that seems more like mainstream treatment.
But because I have a long history of doing this regional chemotherapy
work, I really can envisage us developing strategies for giving
the viruses locally — peritoneal cavity, pleural cavity —
for patients with recurrent effusions, et cetera, et cetera.
DR. LOVE: I like the concept
of a stealth virus.
DR KERR: Yes. Yes.
DR. LOVE: What about potential
toxicities? Have you seen anything or are you concerned or worried
about anything?
DR KERR: Gene therapy is one of the most highly
regulated and complex health areas in the world. And, therefore,
we’re very cautious with it. But in our Phase I trial, giving
the virus directly into the liver tumor nodules, no toxicity at
all. And in combination with the prodrug, once the prodrug is activated
to the cytotoxic species, there’s a half-life of only seconds.
Therefore, it’s locked into the cell, doesn’t diffuse
out into the bloodstream, and, thus would be very unlikely to cause
the sort of systemic toxicity that we see with conventional chemotherapy
drugs.
DR. LOVE: How confident are
you of the specificity of this virus for tumor cells?
DR KERR: There are a number of ways that we can
build specificity in. One is we can alter the tropism of the virus.
By that, I mean, we understand how the virus binds to its cells,
and we can change that so that we can increase the potential for
the virus to bind to cancer cells, as distinct from normal host
cells. So, that’s an area that we’re actively researching.
The second is in terms of molecular switches. So, we could constrain
the virus so that it can only divide in cells that have got mutant
P-53, for example. That’s a good one. Or we’ve done
some very nice work with things called promoters, which are on-off
switches for DNA. So, if we wanted to make sure that our enzyme
gene was only switched on in colorectal cancer cells, we could use
the CEA promoter, which is switched on in 85 percent of colorectal
cancer cells and switched off in the vast majority of normal human
cells. So, by putting the CEA promoter upstream of, and therefore
controlling the enzyme gene NTR, we could get specificity that way.
And we’ve done experiments showing that.
DR. LOVE: Any other biologic
approaches that you’re excited about or that you think five-ten
years from now are going to be translated to clinical practice?
DR KERR: Because I’m particularly interested,
academically, in adjuvant chemotherapy, and since moving to Oxford,
which has got one of the best immunology centers in the world, I
think we’ve got to keep our eye on immunotherapy. I think
cancer vaccines — I think their day will come again. I mean,
all of those tunes of the ‘80s and ‘90s — we’re
used to the IL 2s and the interferons and the promise that was held
and subsequent disappointment. Then, of course, eventually, they
find some true place in the therapeutic armamentarium.
But, there’s such extraordinary insights into the molecular
understanding of immunology, how tumor antigens are presented and
so on, that these new generations of vaccines, whether peptide vaccines,
whether given in viruses, I think it’s very exciting. I’m
itching to get my hands on some of these immunotherapeutic vaccines,
so that we can factorialize them into our adjuvant trial design.
Again, I think over the next ten years, we in the community should
test these agents properly, not in advanced disease. It doesn’t
make sense. But let’s take them into an adjuvant setting and
let’s give them the best chance to return some health gain.
DR. LOVE: Biologically, I certainly
understand why it makes more sense to go directly into the adjuvant
setting, but is that really going to be realistic? We’ve always
had this thing that we have to test agents in the metastatic setting
first. Do you think that the research community is ready for using
things directly in the adjuvant setting?
DR KERR: I am absolutely convinced and have made
a commitment to have our adjuvant group to test these novel vaccines.
So, if you can assure me of safety, not efficacy, but safety, and
if you can demonstrate to me that they’re truly immunogenic,
that we get cytotoxic T lymphocyte response, the clever immunology,
then I would take that novel immunogenic agent into a randomized
trial in the adjuvant setting without any doubt or question at all.
I’d probably factorialize it. So if we’re asking a big
chemotherapy question, A versus B, then I’d run a two-by-two
factorial trial with the vaccine coming in, as well.
DR. LOVE: Any reason to think
there could be any antagonism if you had chemotherapy on board,
or would you wait until the chemotherapy is completed? DR KERR: It’s a very interesting question,
as to how conventional cytotoxic drugs may or may not be immunosuppressive,
and how that may reduce the efficacy of a vaccine. Actually, there’s
not a huge amount of work done in that. There’s very little
work done in 5-FU. If anything, 5-FU is mildly immunostimulatory.
It stimulates NK cells, which are killer cells, rather than inhibiting,
and, therefore, I think we need to do the studies. So, the bits
of information that I would want before taking a novel vaccine into
an adjuvant setting are safety, immunological efficacy and, if I
were going to give it in combination, I would want to know that
the combination didn’t immunosuppress or cause problems that
way. But it would be reasonable to design a trial in which you gave
chemotherapy, finished it, and then randomized to the vaccination
after that. There’d be no problem with that, really.
|
