Capecitabine Research Continues to Shed Light on New Combinations of Agents to Treat Breast and Other Solid Tumors

CHICAGO-Capecitabine(Xeloda) research continues to bring tolight new combinations of agents in thetreatment of metastatic breast cancer, andto test the combination of capecitabine/radiation therapy in the treatment of othersolid tumors. A series of symposia sponsoredby Roche and held in conjunctionwith the 39^th Annual Meeting of theAmerican Society of Clinical Oncology exploredthe use of capecitabine, a tumoractivatedoral fluoropyrimidine, in varioustreatment regimens.Essence of RadiosensitizationEdgar Ben-Josef, MD, of The KarmanosCancer Institute and Wayne StateUniversity in Detroit, explained howradiosensitization can be used to controlgastrointestinal (GI) malignancies withoutincreasing the complication rate. "Thisis the essence of radiosensitization. Wewant to achieve a therapeutic gain. And,for effective radiosensitization, I can'toveremphasize the importance of not increasingtoxicity," Dr. Ben-Josef said.Forty years ago, early work withfluorouracil (5-FU) showed that combiningineffective chemotherapeutic dosesand ineffective radiation doses could resultin cures in 30% of tumors in animalmodels of cancer. Subsequent researchrevealed the importance of correct timing.For example, to be effective, 5-FUmust be in the tissues within 8 hours afterradiotherapy, Dr. Ben-Josef explained.Dr. Ben-Josef reported his findingscomparing 5-FU and capecitabine andtesting a capecitabine/radiotherapy combinationin prostate cancer cell lines."5-FU adds nothing to radiation," he said,"but capecitabine adds an important synergisticeffect."To summarize the phase III data oncapecitabine vs 5-FU in metastatic coloncancer, the response rate with capecitabinewas 25% compared to 16% with 5-FU.The time to progression for both agentswas equivalent, as was survival. But accordingto Dr. Ben-Josef, "there are clearlyfewer grade 3/4 toxicities with capecitabine."Advantage in Lowering DosesA study among 32 patients with a varietyof malignancies in the GI tract foundthat the capecitabine/radiotherapy combinationhad an advantage in older patients.The median age of the patients inthe study was 67 years, but patients up to84 years were included. Dr. Ben-Josef said,"Because this combination of capecitabineand radiotherapy is well tolerated, youcan treat patients you might not otherwiseconsider for aggressive treatment,either because they have comorbidities ordue to their advanced age."The treatment regimen started withcapecitabine 1,250 mg/m² bid, but it wasreduced to 625 mg/m² bid. The mediandose, and the one most commonly used,was 800 mg/m² bid, 5 days a week withradiotherapy. In a phase I study, therewere no dose-limiting toxicities at thisdose.Results from the laboratory indicatethat with capecitabine there is effectiveradiosensitization at much lower dosesthan the maximum tolerated dose (MTD)."This is the situation that we face withmost chemotherapeutic agents today," Dr.Ben-Josef said. "For substantial tumorresponses, we have to accept fairly highcomplication rates, and therefore the paradigmhas been to push the dose to theMTD, to operate just below the MTD. Itdoesn't make sense now to work in theregion of the MTD. The therapeutic gainis much higher at lower doses. So I submitthat with newer, tumor-selective drugswe need to start working with a newparadigm: to identify the minimum effectivedose (MED), then operate a notchabove this minimum. Then we can explorethe full potential of the drug beinginvestigated."Metastatic Pancreatic CancerCurrent trials using capecitabine totreat inoperable and metastatic pancreaticcancer were reviewed by Robert Fine,MD, Columbia University, New York.Pancreatic cancer has a poor responserate to chemotherapy (ASCO abstracts1129 and 1517). "The mechanisms of drugresistance are very important for understandingwhy these tumors are so resistant,"Dr. Fine said."First, to kill these cells, we have toexploit biochemical synergisms, and labdata show that if you achieve a biochemicalsynergy between drugs being used,you can overcome some mechanisms ofdrug resistance. Second, with synergisticchemotherapy you can significantly lowerthe chemotherapy concentrations andstill kill the same number of cells-moreis not always better. So you add finesse towhat the oncologist wants to achieve:shrink the tumor without hurting thepatient. Third, you want to target pathwaysfor p53 independent apoptosis because75% to 80% of all pancreatic cancertumors have mutant p53-tumor cells inpancreatic cancer patients are not goingto die through the classic p53-dependentapoptotic pathways."The Finesse of OncologyEchoing Dr. Ben-Josef's descriptionof the importance of synergism in combinationtherapies, Dr. Fine said, "You wantto reduce doses without sacrificing efficacy-this is what I call the finesse ofoncology.Dr. Fine's goal in treating pancreaticcancer is to induce p53-independent apoptosis."The gist of our work is that wefound in human pancreatic cancer celllines, gemcitabine (Gemzar), docetaxel,and 5-FU produce synergistic cell-kill.When we used this synergism, we wereable to reduce the doses by 50% and stillkill the same number of tumor cells.""And for simplicity of administration,we used capecitabine. We gave 42 metastaticpatients 750 mg/m² of capecitabinetwice daily. The result was a completeresponse rate of 9.5% and a partial responserate of 47%. Also, 74% had a clinicalbenefit response," Dr. Fine reported.GTX Regimen and Variation"We used an apoptotic assay with ahuman pancreatic cancer cell line. Wefound that 5-FU and gemcitabine killed asmall number of tumor cells, and docetaxelkills a few more. But if you put themtogether, you get this powerful synergism.That experiment led to the regimen thatwe refer to as GTX."The GT-X regimen consists of capecitabine500 to 750 mg/m² twice daily for 14days. After 4 days, patients receive gemcitibine750 mg/m² over 75 minutes andlow-dose docetaxel (Taxotere) 30 mg/m²over 30 minutes. It is a 2-week regimenwith the third week off, and responses areevaluated after three cycles.In the case of GTX failure, Dr. Finethen separated the administration of thedrugs. This was based on lab experimentsshowing resistance to GTX was due tocytokinetic cell cycle resistance. The samethree drugs that had just failed were usedagain, but this time in a sequential manner.This modified regimen is referred toas sequential GT-X. Dr. Fine comments,"This is a very easy treatment for the patient.The dose is about 60% to 75% of theGTX doses. In a preliminary study of 10patients who failed GTX, there was onecomplete response and two partial responsesof their metastatic disease. Fourother patients achieved stable disease.Median duration of stable disease was 5.5months. Three patients did not respond.There were no grade 2 or higher toxicities,"he explained."The GTX regimen is based on in vitro synergisms and equal activity in mutantand wild type p53 PC cells," Dr. Fine said."The toxicity profile is favorable. Lowerdoses still kill tumor cells. Response ratesare promising in the advanced state, andfurther studies are warranted to ascertainthe true clinical benefit," Dr. Fine concluded.Metastatic Breast CancerA phase III study of capecitabine plusbevacizumab (Avastin) in metastaticbreast cancer was discussed by Kathy Miller,MD, of Indiana University MedicalCenter, Indianapolis (ASCO abstract 766)."For the past 2 to 3 decades, it has beenknown that tumor growth is dependent ofangiogenesis. This discovery generated aninvestigation as to what factors stimulateor inhibit blood vessel growth. Vascularendothelial growth factor (VEGF) is oneof the most potent stimulators of angiogenesis,and this finding led to explosivegrowth in the development of agents thatinhibit VEGF signaling. One of the earliestagents was bevacizumab, a humanizedmonoclonal antibody directed against VEGF itself. It recognizes all of the VEGFisoforms, and it has activity in monotherapy,"Dr. Miller explained.In heavily pretreated breast cancerpatients, about 9% of patients have anobjective response to bevacizumab monotherapy,Dr. Miller reported, and an additional10% are without progression at 22weeks. This finding led to a phase III trialfor patients with prior anthracycline ortaxane therapy for metastatic breast cancer,or who had a relapse within 12 monthsof completing anthracycline and taxanecontainingadjuvant therapy.The study had 462 patients who werewell balanced for demographics and all ofthe prognostic factors that might influenceprogression-free survival and overallsurvival. Three-quarters of the patients had visceral disease, so it was a difficultpatient population. Both capecitabine andbevacizumab toxicity mirrored what wasseen in earlier trials.Response Was IncreasedBut Not the DurationResponse rates, as evaluated by investigatorsor by a blinded review facility,nearly doubled, but there was no differencein duration of response. Thereforethe progression-free survival data are notmuch different between the two treatmentarms (4.17% for capecitabine, 4.86%for capecitabine plus bevacizumab). Therewas no improvement in median progression-free survival."When we measure median progressionfree survival, we measure the median,"Dr. Miller stressed. "Response ratesof 10% to 20% do not drive median progression-free survival. That is driven bythe 80% to 90% of patients who did notrespond to therapy. Time to responseshows the time the patient is in responseto therapy. If the patient never responded,the patient has a time of 0. In thisstudy, at each time point, the patients whoare given the combination therapy aredoing slightly better, but most of thatbenefit is early in the treatment. By thethird time point, most of the excess responsesgained by the combination armare lost. So, there are excesses, but they areshort-lived. And without that, you willnever see an improvement in the time toprogression," Dr. Miller explained.VEGF Expression Matters"VEGF expression does matter, butprobably not in the way that most peoplewould have predicted, " Dr. Miller said."Basically, this was a negative studybecause it was powered to look at differencesin median progression-free survival,and there were no differences, but inmy mind it is a critical proof of conceptstudy, and it really validates the importanceof antiangiogenic therapy. That conclusionis because of the clear increase inobjective response rate while taking intoaccount that this was a very difficult patientpopulation," she continued."Our challenge now is to take advantageof that activity so we can get evengreater benefit for our patients," Dr. Millerconcluded.