New Tyrosine Kinase Inhibitor More Potent Than STI-571

MIAMI BEACH -The Bcr-Abl tyrosine kinase inhibitorPD173955 (PD17) binds to the targetATP binding pocket even moreefficiently than STI-571 (imatinibmesylate, Gleevec). It shows 15 to20 times greater efficacy in chronicmyelogenous leukemia (CML) celllines because it can bind to eitheropen or closed activation loops.Bayard D. Clarkson, MD, presentedthe research at the Molecular Targetsand Cancer Therapeutics meeting,sponsored by the AmericanAssociation for Cancer Research, NationalCancer Institute, and Europe-an Organization for Research andTreatment of Cancer (abstract 568)."We basically pulled this compoundout of the literature, synthesizedit, and found that it worked,"said Dr. Clarkson, Enid A. HauptChair of Therapeutic Research andhead of the Hematopoietic Cell KineticsLaboratory at MemorialSloan-Kettering Cancer Center. TheMSKCC researchers collaboratedwith Dr. John Kuriyan and otherRockefeller University investigatorson the x-ray crystal structure work.The initial translocation betweenchromosomes 9 and 22 in CML producesa BCR-encoded sequence fusedto a truncated c-Abl gene. The Bcr-Abl protein encoded by this mutationgreatly increases c-Abl's tyrosinekinase activity and causes the clinicalmanifestations of CML. STI-517 producescomplete hematologic responsesin most cases of CML because itbinds to an ATP binding pocket andblocks this process.Dr. Clarkson's team set out tofind inhibitors of Bcr-Abl that mightbe more effective than STI-571,which is met by resistance in patientswhose CML is in blastic phase.PD17 is more effective than STI-571 at inhibiting cell lines containingBcr-Abl in vivo and at blockingCML progenitor cells without inhibitingnormal progenitor cells. Invitro assays showed that PD17 inhibitedfresh CML primary progenitorcells in the low nanomolarrange. Dr. Clarkson said that PD17is active and tolerable in mice at levelsthat should be active in humans.He presented x-ray crystal structuresof the catalytic domain ofc-Abl (Abl kinase) complexed witheither STI-571 or PD17. Theseshowed that STI-571 can bind tothe target site only when the 21-residue activation loop of c-Abl isin a closed configuration that resemblessubstrate binding. PD17binds when the activation loop isopen and resembles that of an activekinase."Modeling shows that PD17 canprobably target c-Abl regardless ofthe activation state it is in, but STI-571 can only recognize the uniquedownregulated form. The greaterpotency of PD17 is probably due, atleast in part, to its ability to recognizemultiple states of the kinase.We're not sure that is the entire reason,however," Dr. Clarkson said.Knowledge of the three-dimensionalstructure of the Abl kinasedomain and how the specific inhibitorsdiffer in their binding propertiesat a molecular level should allowscientists to make rationalmodifications to design and produceeven more potent and specific drugsfor treatment of leukemia and othertypes of cancer, he said."I think that these tumors aregoing to develop resistance even tothis drug [PD17], so we need morethan one molecular target. I thinkthat if we had three targets, as we doin promyelocytic leukemia, there isa good possibility that we could curethese tumors," he said.Dr. Clarkson also said that hisgroup has studied anotherParkeDavis (Pfizer) compound,PD166326, that is about fourfoldmore potent than PD17 in inhibitingAbl kinase. It is now being studiedin CML mouse models."PD17 and PD16 are also muchmore active than STI-571 against a varietyof human glioblastoma, sarcoma,and neuroblastoma cell lines, buthigher concentrations are needed thanin the case of CML cells expressingBcr-Abl," Dr. Clarkson said.