Perspectives on Fluorinated Pyrimidine Use in Gastrointestinal and Breast Oncology

During this symposium, facultyparticipated in a roundtable discussion of the relative merits and weaknessesassociated with use of the fluorinated pyrimidines. Dehydropyrimidinedehydrogenase (DPD) deficiency and the impact of this deficiency on the efficacyof fluorouracil-based therapies was a major focus of the discussion. Othertopics discussed included appropriate follow-up for patients receiving oraltherapy and reimbursement issues associated with oral vs intravenous therapy.

MODERATOR:

Leonard Saltz, MD
Memorial Sloan-Kettering Cancer Center, New York, NY

PANELISTS:

Al Benson, III, MD
Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, Illinois

Peter Danenberg, MD
University of Southern California, Los Angeles, California

Robert Diasio, MD
University of Alabama Comprehensive Cancer Center, Birmingham, Alabama

Jean Grem, MD*
National Cancer Institute - Medical Branch, National Naval Medical Center, Bethesda, Maryland

William Gradishar, MD
Robert H. Lurie Comprehensive Cancer Center of Northwestern University, Chicago, Illinois

Daniel Haller, MD
University of Pennsylvania, Philadelphia, Pennsylvania

Paulo Hoff, MD
The University of Texas M. D. Anderson Cancer Center, Houston, Texas

John Marshall, MD
Georgetown University, Washington, DC

Udo Vanhoefer, MD, PhD
University of Essen Medical School, Essen, Germany

Dehydropyrimidine Dehydrogenase Deficiency

Leonard Saltz, MD: In this general discussion, we’ll talkabout the direction in which we see oral fluorinated pyrimidines going. Theissues there are going to be in terms of their development.

Let’s go back to the creation. It’s day six. Human beingsare being created. You’re putting the person together and you throw in DPD.Why? It’s fascinating to me that people who are DPD deficient live all theirlives without knowing it unless they are exposed to fluorouracil (5-FU).Eniluracil is a phenomenally effective inhibitor and yet unbelievably nontoxic.So what is this enzyme doing there?

Robert Diasio, MD: We talk about DPD and fluorouracil metabolismas being a catabolic enzyme. But in terms of metabolism itself within the body,we can also look upon DPD as being an anabolic enzyme and a synthetic enzyme inthat it is one of the enzymes responsible, eventually, for the synthesis of betaalanine. Beta alanine has been suggested, by some people, at least, to bepotentially important as a neurogrowth factor, although the evidence is veryweak.

But there’s another symptom of DPD deficiency that’s beendescribed in children that results in growth and mental retardation. A number ofcases have been described, mainly in the Benelux countries (Belgium, TheNetherlands, and Luxembourg), where they happen to do amino acid analyses on allnewborns. They were the first to describe DPD deficiency. They discovered anelevated peak and couldn’t explain what it was; it turned out to be uracil.They found this to be associated with microcephaly, growth retardation,developmental defects, and seizures in some of these children. So beta alaninemay have a role in terms of development. That’s why I would ask why DPD isneeded. It wasn’t put there just to cannibalize 5-FU.

Peter O’Dwyer, MD: What about other altered pyrimidines likemethyl uridine that are important metabolically, for which you still wouldn’twant very high levels?

Dr. Diasio: It’s not clear. There’s very little informationabout the role of these other pyrimidine compounds.

Jean Grem, MD: It’s even more complicated because there havebeen cases described where there are two siblings who are both genotypically DPDdeficient and yet one of them has all these neurologic deficits and the otherone is perfectly normal. So it’s more complicated than simply asking why youhave DPD deficiency. For people who are absolutely deficient, it’s possiblethat something else may be missing.

Dr. Diasio: At the molecular level, some of those mutations havenow been shown to be the same. Skipping of axon 14 actually has been shown to beresponsible. We’ve shown it in some of the 5-FU patients.

Leonard Saltz, MD: So patients who find out that they are DPDdeficient when they have a bad experience with 5-FU, presumably have a lesserdegree of dysfunction—but some function—because they don’t havedevelopmental abnormalities. Phenotypically, is there anything to alertsomebody?

Dr. Diasio: No. It’s a true pharmacogenetic syndrome.

Dr. Saltz: If it’s really pharmacogenetic, then it’s harderto justify the existence of the enzyme in the first place. It may be somethingyou need early on and then it’s superfluous.

Dr. Diasio: Yes. It may be very important in fetal development.

Dr. Saltz: Except if that were true, we’d have to postulatethat the people who clinically had deficiencies have lost them somewhere alongthe way; that they had to have had it to develop normally.

Dr. Diasio: Or they had an exogenous supply of beta alanine orsome of these other factors.

Dr. Grem: For these children there may be some in utero deficitor dietary deprivation. All that isn’t really known. DPD protein is a hugemolecule and very complex, 33 iron atoms. Its substrates are these tinypyrimidine bases and it doesn’t seem logical that you would have this hugeenzyme to handle these little bases.

John Marshall, MD: Is it inducible? Can you change itsexpression?

Dr. Diasio: Not really. Not in the same way as a number of otherenzymes that we traditionally think of as being inducible.

Dr. Saltz: We don’t understand the regulation of itsexpression.

Peter Danenberg, MD: It’s likely that this enzyme also mighthave some undiscovered role.

Dr. Diasio: We’ve looked at the homology to see if there’sanything else. We wondered if it was related to other enzymes. It doesn’t seemto share homology with a lot of these other enzymes. Going to the gene bank, wehaven’t been able to find a similarity.

DPD Deficiency andResponse Rate

Dr. Saltz: What’s interesting also is that when we look at atumor and talk about the tumor as being a high expressor or a low expressor ofDPD, that doesn’t necessarily correlate in any way with the patient. There’sthe phenotype that the patient has in terms of tolerability of 5-FU and thenthere is the tumor activity of the enzyme that may have prognostic importance.

Dr. Diasio: That’s one of the things we’re actually lookingat in the patients that we study. We’re also looking at the tumor DPD activityand peripheral blood mononuclear (PBM) cells and they don’t seem to agree.

Dr. Grem: If you look at the tumors in DPD-deficient patients,the tumors may also be DPD deficient.

Dr. Saltz: But the low DPD tumors, according to Kathy and PeterDanenberg’s data, are more likely to be sensitive to 5-FU and the high DPDsare not sensitive to 5-FU. So it’s interesting that all of these inhibitors,whether they’re competitive or noncompetitive, don’t really seem toinfluence a substantial difference in the response rate.

Dr. Diasio: Where we’ve looked at our patients, we haven’tseen any real difference in terms of tumor response. In DPD-deficient patientswho’ve been treated with 5-FU, we haven’t seen a better response.

Role of Eniluracil

Dr. Marshall: The extension of that is what you would expect,because if eniluracil inhibits DPD in tumors, you would expect a markedly higherresponse.

Dr. Saltz: That was certainly the hope in the development ofeniluracil, but now it looks like we’ve stepped back from that.

Dr. Diasio: It was mentioned that eniluracil itself isrelatively nontoxic, but there are some severe toxicities. I think about 18patients have been described who have had severe toxicity after they receivedthe second dose of 5-FU.

Dr. Saltz: That’s toxicity with 5-FU. If you give eniluracilby itself, are there toxicities with that—if you don’t expose somebody to afluorinated pyrimidine?

Dr. Grem: The only experience we have is a 7-day course ofeniluracil by itself and that’s it, so you don’t know about longer-termtherapy. The ongoing phase II and phase III trials aren’t reporting anythingreally unusual happening but there has been some neurotoxicity in a phase Itrial with twice-daily dosing over 28 days. They did have one patient who hadbeen on study for quite some time who developed cerebellar atrophy, withabnormalities on MRI scan.

In our current trial with eniluracil, 5-FU, and leucovorin, wehad two patients develop peripheral sensory motor neuropathy. That’s somethingyou wouldn’t expect. Patients who are DPD deficient and who can’t make anycatabolites clearly can have acute neurologic toxicity that can progress tocoma. On the other hand, there is some preclinical data that suggest that somecatabolites may directly harm brain tissue or neurofibers.

Dr. Saltz: Really, what we’re talking about is 5-FU toxicityin the setting of DPD depletion. Depletion of DPD itself doesn’t seem to hurtyou. It seems to be an enzyme that’s along for the ride and seems to exist tofrustrate oncologists.

Dr. Diasio: Only in adult life does that seem to be true.

Dr. Saltz: So we’d have to be very concerned about the use ofthis in pediatrics.

Dr. O’Dwyer, MD: In families that have relative DPDdeficiencies, is there any evidence of a higher instance of cancer or of anyother disease?

Dr. Diasio: No. We’ve looked for that and we have noassociated abnormalities. We haven’t seen any large chromosomal defectseither. All of the mutations that have been looked at have been found to bemainly either axon-skipping or just substitutions.

Dr. O’Dwyer: What about molecular abnormalities in the colontumors?

Dr. Diasio: We really haven’t characterized them, other thanthe messenger RNA expression. Patients who are completely deficient are veryrare.

Udo Vanhoefer, MD: Are there any standards defined for measuringDPD and thymidylate synthase (TS)?

Dr. Diasio: Yes. We pretty much use the same methodologies.

Dr. Grem: Something to think about with eniluracil is that withthe current schedules of many of the infusional regimens, the tolerated dailydose of 5-FU is more than a hundredfold lower than what you would give to beginwith, based on the preclinical animal models and animal allometricpharmacokinetic scaling. It’s assumed that renal clearance is about 10% of5-FU clearance, so the tolerated dose would be 10% lower, but in fact, with theschedules that mimic infusional schedules, they’re more like a hundred foldlower. It leads you to think that what we may actually be doing is just allowingmuch, much greater intracellular accumulation of 5-FU metabolites and maybe someof the NMR investigators and the PET investigators will be able to documentthat. But it stands to reason that the toxicity and the tolerated doses are muchgreater than you would predict based simply on an effect due to changes inclearance.

Dr. Saltz: So where do we go with this? Which of these drugs isgoing to be the one to work on? Do we work on all of them? Do we put themtogether? The issue came up earlier of DPD inhibitors with capecitabine. Isthere a role for that?

Dr. Diasio: I think conceptually it’s an intriguing idea.Eniluracil* is really a very potent DPD inhibitor. It may be more potent thanwe really need and really want. If the tumor data really are true, and we thinkthey are, resistance is due to DPD overexpression in many cases. I thinkeniluracil is exciting as a way of trying to reverse 5-FU resistance due to DPDoverexpression.

Dr. Saltz: But shouldn’t we have seen that already?

Dr. Diasio: Well, it’s disappointing in situations, such asEllen (Hollywood) mentioned about hepatoma, which was one of the first trials.We know that hepatomas are loaded with DPD. In fact, they have higher DPD levelsthan the liver, which is the main site of DPD. It was very disappointing to seethat the hepatoma data really haven’t shown much benefit for eniluracil/5-FU.

Al Benson, III, MD: It’s hard to imagine just one molecularevent that will determine outcome of metastatic disease. Perhaps it’s a hostof events, but we hope it’s not so individualized that you have to look at 20or 50 different markers.

Paulo Hoff, MD: All the other fluoropyrimidines are going toface, at least in colon cancer, the same question that UFT (uracil and tegafur)faced. They are all being evaluated in phase III trials, which essentially aredesigned to show equivalency.

Dr. Grem: I think there is compelling data demonstrating thecontribution of eniluracil to the 5-FU regimen.

Daniel Haller, MD: And now you’re changing it from $5 worth of5-FU to 15 cents worth in the oral form.

Dr. Saltz: Percentage-wise, it’s a great savings.

Dr. Haller: But is there a compelling argument why any one ofthe drugs, other than being first in line, should be the only approvable drug?That’s really the issue.

Capecitabine

Dr. Hoff: Obviously capecitabine (Xeloda) gives you a higherresponse rate. But the survival is the same. UFT got the same survival and Iwould suspect eniluracil is going to get the same survival.

Dr. Haller: Is there one that’s clearly less toxic than anyother?

Dr. Hoff: Depends on what you call toxicity, I guess. Whattoxicities are you interested in?

Dr. Haller: That’s the only selling point if they’reequivalent in efficacy, unless you could do studies of eniluracil in patients toshow that people who had high DPD deficiency actually were the ones whobenefited. You could then pretest and treat only those patients witheniluracil/5-FU.

Dr. Diasio: The strongest point that has to be dealt with is thesurvival. Speaking for eniluracil vs uracil as the inhibitor, the pharmacologyis absolutely much better with eniluracil but ultimately the real question isgoing to be the survival data. The FDA has made that very clear.

Dr. Saltz: One of the problems is, is eniluracil as good as it’sgoing to get? We’re not seeing the therapeutic benefit to the degree that wewould have hoped. We’re seeing a change in the pharmacokinetics, we’reseeing a change in the way we can do it, and we hope that that may translateinto something that is more convenient or better or easier. But let’s face it,the equivalency is always a fallback position when superiority doesn’t seem tobe achievable. Your first goal would be to make it better and you’d be happyto accept even a little tougher toxicity profile if you could clearly show anunequivocal benefit. It’s in the absence of that that we find ourselves in thesecond line of development, which is looking to say: it is as good but it’seasier on the patients or it’s less expensive or somehow better on some softerend point.

Dr. Marshall: One place where a response rate can translate to asurvival benefit is the adjuvant setting. It’s always gnawed at me,particularly with capecitabine, that I envision pockets of metastatic disease orcirculating cells and with IV 5-FU, no activation or limited activation isrequired. But if you had a clone of cells in the adjuvant setting that don’toverexpress thymidine phosphorylase (TP), for example, are you going to getinferior survival benefits in the adjuvant setting? I’ve had patients ask forthe pill for adjuvant therapy. It’s been something I’ve been hesitant to do.Do any of you feel the same way?

Dr. Hoff: The only counter argument is that your liver isnaturally very rich in TP so you’re going to be transforming the drug into5-FU in the liver and hopefully it will circulate. But you’re right. There’s20 times less 5-FU circulating than in the tumor and when you’re talking aboutthe adjuvant setting that may be important.

Dr. Saltz: If we really can characterize via PCR (polymerasechain reaction) that a patient with high levels of TP is going to have betterantitumor activity, then we’re really off to the races.

Irinotecan and Oxaliplatin

Dr. Saltz: To comment on the question about people asking for"the pill" for adjuvant therapy of colon cancer, the answer right nowis there isn’t one. That’s an intriguing experimental question. It’s beinglooked at with a number of different oral agents, but until we have the data, Ithink it’s a big mistake. All too often I find patients asking aboutcombination therapy in the adjuvant setting, be it the incorporation ofoxaliplatin or incorporation of irinotecan (CPT-11 [Camptosar]). My answer isthe same, that the clinical trials are ongoing. We encourage people toparticipate in the trial. They’re rational, ethically designed trials. Butuntil we know the answer, the standard in that setting is 5-FU/leucovorin and Ithink we’d be out on a limb using these different schedules off study.

Dr. Haller: If we’re talking about the oral drugs as beingequivalent to something that’s no longer the gold standard, then the real nextquestion is are any of these drugs more or less combinable with irinotecan oroxaliplatin (Eloxatine) in this country? Irinotecan is a more practical endpoint. Is talking to a patient now about an inferior standard, just because theycan be given more easily, inappropriate? If you hold to that as being easier, isthere attractiveness for any of the three likely available drugs, more or less?Because that would pull one of them ahead of the pack.

Dr. Hoff: One thing that I think we’re going to have toeventually consider is what the Japanese have been doing, using the same dosefor everybody. There’s no compelling data that tells us that we should baseuse on body surface area. If you had a single dose or at least two schedules—dependingon if you’re more than 1.5 or less than 1.5 meters square—it could simplifythe tablets. In practice, I think nobody uses 150-mg tablets. You just round up.

Dr. Saltz: Probably one of the less productive exercises we gothrough is dosing irinotecan on a per meter squared basis when you consider it’sa prodrug that’s activated to different degrees and then deactivated todifferent degrees and potentially influenced by various other drugs and so on.In reality, we could probably take a round number to start and have a similarnumber of dose adjustments based on toxicity. That argument perhaps could bemade for any of the oral agents. One would think that if eniluracil is as muchof a leveler of the playing field as it appears to be in terms of themetabolism, it might be easier to make a uniform dose of 5-FU. But that’s purespeculation. It’s hard to say which of these drugs has a leg up on the otherin terms of combination.

Dr. Benson: It’s going to be a real dilemma, because right nowthe numbers of patients who actually go on trial—advanced disease trials, andcertainly adjuvant trials—are tremendously resource heavy. We’ll have quitea substantial collection of phase II trials and phase I trials with multiplecombinations and I suspect they’ll all be fairly similar. Is it worth theresources to pursue all this? If you just look at the combinations withirinotecan or oxaliplatin alone, there would be multiple trials, which right nowwouldn’t be doable.

Dr. Hoff: What’s going to happen is that capecitabine is goingto take the lead, at least in the United States. The other combinations would beused more in Europe and South America, where the other drugs are available. Butright now, I think capecitabine is going to take the lead because it’s what’savailable.

Dr. Saltz: There’s a lot of pragmatism to your statement. Thequestion then is how should we go forward in terms of combinations? What are theadvantages of a capecitabine/irinotecan combination vs a 5-FU/irinotecancombination?

Dr. Vanhoefer: In Europe, we have to use the portable devicesfor infusional 5-FU. So for Europe, it’s a huge advantage to use capecitabine.I think the next step is to replace infusional 5-FU with oral capecitabine andif it works with oxaliplatin or with irinotecan it would be fine. Trials areongoing with the combination in the adjuvant setting, and if we have an adjuvantsetting advantage for this combination, that would be a major advantage becausewe would not have to implant ports.

Dr. Hoff: There’s a potential interaction between thetopoisomerase I inhibitors and resistance to fluorinated agents, and by using aprotracted dosing, in this case capecitabine, or whatever oral agent you preferto use, you actually increase your chance of having an efficacious result over5-FU as a bolus.

Dr. Vanhoefer: Second point, I think, is breast cancer. Thetaxanes and anthracyclines are going more and more to frontline, so if you cancombine capecitabine and taxanes it might be a very nice approach. The data forthis infusional approach are very active, about 40% to 45%.

Oral vs Intravenous Therapy

Dr. Saltz: These suppositions are based on the assumption thatpatients prefer the pills. Do we know that? When you ask somebody would yourather take pills twice a day or come into the doctor’s office and get aneedle stuck in your arm, it’s not a very difficult question. But when youreally ask patients what they think about oral chemotherapy, their expectationsare that it’s going to be essentially nontoxic, that it’s going to be liketaking a vitamin tablet or a Tylenol. And people are sometimes very disappointedto find out that these drugs have side effects and sometimes the same sideeffects that we’ve come to anticipate with parenteral drugs. When you reallybreak it down and explain to a patient that we have to worry about hand-footsyndrome, we have to worry about diarrhea, we may have to keep an eye on yourblood counts, we may have to watch your liver chemistries, this may interactwith lots of other drugs that you’re taking, how great a degree of acceptanceover the infusional pump are we likely to see?

Dr. Marshall: I have personal experiences that patients don’twant to go back to the IV. When you put them on an oral agent and then theyprogress and you have to go back to an IV, they ask can’t we just try thepills for a little while longer? Even when they’ve got hand-foot syndrome andthe like.

Dr. Hoff: Patients prefer the oral form.

Dr. Haller: Hardly anyone turns down pump therapy because they’reencumbered by it. You still have a reason to bring them into the office fortheir IV.

Dr. Marshall: That solves part of the reimbursement issuebecause you’ve still got the patient in the chair for 2 hours.

Dr. Haller: Perhaps the best thing that has happened to the oralmedications is irinotecan.

Dr. Benson: A big factor is what keeps the patient home. In oursystem, we use visiting nurses who handle the pump, so that’s essentiallyequivalent to oral therapy because someone comes to their home. They are notgoing to the office.

Reimbursement Issues

Dr. Saltz: Of course, the interesting problem that will nowarise from a third-party payer point of view is the doctors and the patientswill be happy with the combination of oral chemotherapy and parenteralchemotherapy because we need the combination therapy to maximize efficacy.Secondly, it gives us a chance to see the patient, and educate the patient,evaluate the patient, and it’s what we all think is safer care at this point.I don’t think we’re ready to have the patient home long-term as we areperhaps with the AIO. But when we figure out at least thedirect medical costs there, it’s going to be higher because you’re going tohave the increased drug costs of the newer agent and you’re going to have thecosts of paying for physician supervision and nursing supervision and thetechnical charges and so on. Now that’s not necessarily a bad thing in termsof just saying you’re going to give better care and it’s going to cost more.That just may be the reality. But to try to give better care at the same priceis something that I think third party payers have been looking for for a longtime, and they keep running into a wall.

Dr. Haller: So what would the drug cost be for a 6-week courseof irinotecan/capecitabine?

Dr. Saltz: The direct medical cost of giving the irinotecan, aswe figured it out, was about $22,000 dollars for 6 weeks. The capecitabine wouldprobably be about $800. So if you’re an insurance carrier going to combinationcapecitabine/irinotecan, you’ve now increased your drug cost by about $30,000dollars every 6 weeks for a 2.5-month improvement in median survival.

Dr. Haller: If we spend our whole life in clinical researchtrying to do studies to show improvement in survival, and we finally see it, it’snot our problem that it’s expensive. But when you look at newer regimens andit’s $18,000 dollars for a 6-week cycle, it is truly outstandingly high. Sowhen patients come to me and say, you know, your bill was $100 and we have a $10co-pay, they have no idea what the total cost of their drugs are.

Dr. Saltz: I think it’s important to realize also that some ofthose costs we talk about for parenteral use are in fact the costs that arepaying the doctors and the nurses, and they are providing many services that arenot specifically reimbursable otherwise. If you take that payment away and stillexpect those services to be provided, it may not be realistic. What you’regoing to have is a drop in the safety profile of the oral agents. One of thethings that I’ve heard a general consensus about in terms of the discussionsearlier this morning is that you can’t expect to treat somebody with an oralchemotherapy by writing a prescription and say "See you in 6 weeks".

At least for the first few cycles, until people are on a steadydose and we know how they’re going to tolerate it, they need to be seen fairlyregularly and they need to have significant nursing intervention. So if we lookat trials where they had that intervention and they had that nursing follow-upand they had the nursing phone calls, and then we try to extrapolate that to nowgo out and just give the drug to somebody by mail order, that may be just asinappropriate as looking at dose at one level and trying to then use it atanother. We don’t really know how that’s going to work out. We have todevelop something that is truly exportable to the real world of practice and ifwhat we’ve developed is something that works with expensive, high qualityresearch nursing and research support staff, we haven’t fully evaluated it inthe community.

Dose Schedule vs Cost

Dr. Saltz: Let me go back to the dose question for a minute,because we did discuss that we’re not real comfortable with capecitabine at2,500 mg/m², and it’s certainly been my limited experience with the drug thathand-foot syndrome is a significant concern at that level. I would argue if wereally want to believe the data, that there’s an intratumoral activation ofthis drug. We should be able to use it at a relatively nontoxic systemic doseand have substantial doses in the tumor. Is that a practical thing to do? Are wegoing to see a study in colon cancer or in breast cancer where we look at 2,000mg/m² vs 2,500 mg/m²?

Dr. Hoff: I can tell you that it seems to work. I do not know ifit is the same, but we use it quite often. We use it off-schedule andempirically. It does work. It’s not a large trial, but to do a large trial isalmost impossible.

Dr. Marshall: I actually think capecitabine is a different drug.It’s not just infusion of 5-FU. The net result is delivery of 5-FU, but Ithink the response rates bear that out. So what I would argue for would be anice way to shorten the schedule and increase the dose. Manipulate schedules.Try to up the dose response curve a little bit with 5-FU. Can you influence theupregulation of TS, for example, by changing the schedule? Might higher, harderhits less often bring costs down, and simplify schedules? There are a lot ofthings that might push the response higher, which would then be an interestingthing to bring to randomized study.

Dr. Saltz: To some degree, it’s almost theoretical because nowyou’re basically going for bolus oral regimens.

Dr. Marshall: Right. But delivering a dose that’s nowacceptable. I mean you can’t get to those levels of bolus, but you mightachieve a super high-dose 24-hour schedule.

Dr. Haller: Has anybody done that with Capecitabine?

Dr. Grem: Initial studies with capecitabine used either acontinuous schedule or an intermittent schedule. This strategy was a fast-trackclinical development plan and we don’t really know that the intermittent twicedaily for 14 of the 21 days is truly the best schedule. It’s just that itappears to be an effective schedule. With eniluracil/5-FU, the decision has beenmade to go with fixed combination tablets; this has basically restricted theclinical evaluation to a chronic schedule (twice daily dosing for 4 of 5 weeks).If you try to mimic higher-dose, shorter-term infusions, you would end up byasking the patient to take dozens of pills and then you’d be giving themmassive doses of eniluracil, so the pharmaceutical sponsor has sort ofhandcuffed themselves—in the short term at least—with everything banking onthis 28 days out of 35, twice-a-day schedule. From a regulatory standpoint, theMayo bolus 5-FU/leucovorin regimen has been the gold standard; however, manyoncologists in the community use the weekly Roswell Park schedule because theyperceive it as being less toxic and easier to manage for the patients. Plus,after all these years, there are still conflicting camps and regionalpreferences for the best way to give 5-FU. In Spain, they like a weekly 48-hourinfusion. In Germany, they like a weekly 24-hour infusion with high-doseleucovorin.

Dr. Saltz: The argument of what’s the right way to give 5-FUhas gone on for an awfully long time and remains unsettled, and the same thingstarted to happen with irinotecan, although I think a lot of people are agreeingthat it’s just not all that schedule dependent; pretty much any way you giveit, it seems to have a similar efficacy and toxicity profile.

Oral Dosing and Cost

Dr. Grem: Oral irinotecan is in clinical investigation. Most ofthe data with irinotecan is with a short infusion over 30-90 minutes. Whetheryou give it once a week, once every 3 weeks, or once every other week, theefficacy and toxicity are probably about the same. With the more protractedlow-dose daily exposures, it looks like the toxicity, especially the diarrhea,is actually much, much lower. There are no data about drug stability, so if you’redoing protracted infusions of irinotecan, you have to have the patient comeevery single day to the clinic to get the bag changed. But oral irinotecan is indevelopment, and so it may be that some protracted low-dose, 5 days on, 2 daysoff type of schedule may end up being better.

Dr. Haller: Do you think people in practice will really beexcited about having two oral drugs for colon cancer?

Dr. Saltz: I’m fairly certain that we’re not going to see adramatically different efficacy situation. It’s a little too early to tellwhether we’re going to see a difference in the toxicity profile. I think thatwe are interested in developing the oral regimen because there’s going to be arequest for that from patients and whether that’s going to have its problemsin acceptance or not is something we’ll have to deal with as we go. But Ithink it’s something that we should be pursuing.

The Cost of Appropriate Monitoring

One of the issues that we did discuss this morning but neverreally reached consensus on is what kind of monitoring is appropriate. When youstart people on capecitabine outside of a clinical trial, what sort of safetyprecautions do you put in place? How often do you see the patients? How often doyou have your office staff calling?

William Gradishar, MD: I’ll just make a comment about thebreast cancer patients who are off study. Quite frankly, the nursing staff doesn’thave a lot of time, so patients are instructed to call us if they have symptomsor side effects. We see them at least once a month unless they have problems.

Dr. Hoff: Off protocol, we don’t call them every week. We havea pharmacist that reinforces the teaching. We use the forms that have beendescribed and we call patients every 3 weeks when it’s time to recycle.

Dr. Saltz: So once every 3 weeks but with clinical educationearly on?

Dr. Hoff: Yes. And obviously they can call. They have a contactphone number if they have a problem. We really haven’t had any major problem.

Dr. Benson: The nurses go over those sheets carefully. Most ofthe patients, at least our colon cancer patients, have had previous therapy, areaware of the toxicities, and know when to call. But it’s still up-frontnursing time.

Observer: How do you manage the hand-foot syndrome outside ofdose reduction?

Dr. Hoff: There’s no good treatment for hand-foot syndromeright now. People have used everything. Bag balm seems to be a common thingaround the country. People have used nicotine patches. B6 deficiency produced asimilar problem in rats and became the basis for the idea of using B6. We useit, like everybody else. I think it’s an area that needs to be studied in moredepth actually, because we really don’t know even what causes it.

Dr. Saltz: Patients have hand-foot syndrome, people give themB6, and drop their dose and it gets a little better. B6 is cheap and harmless soI suppose there’s not much downside to it but I just can’t really getexcited about it.

Bag balm is very popular. It’s farmer’s ointment for udderinflammation.

Summary

Dr. Saltz: In summary, what we could say is we’ve got twoagents (UFT and capecitabine) that are either here or looking like they might behere soon. One of these is actually on the market (capecitabine) and seems to bethe one that we’re in the situation to do the most work with. We haven’t really been able to identify any particular reason tosay that one of the oral agents is substantially better from an efficacy orsafety point of view than the others, and of course, there haven’t been anyhead-to-head trials, but there don’t appear to be any striking differences ineither the efficacy or safety profiles that we can see.

How we’re going to use these agents is up for grabs. Rightnow, I think that most of the use that we’re seeing for capecitabine is in thesecond-line arena, that it’s being used as a surrogate for infusional 5-FU,and it will be used predominantly in patients who have failed irinotecan. We allhave interest in seeing where these agents will work in combination and thattheir future in first-line therapy is probably in combination withnon-fluorinated pyrimidines such as irinotecan or possibly oxaliplatin. This isan area that is just beginning to be explored and we’ll undoubtedly be seeingmore about it in the next few years.