Document V3GwKa4wO3q0dnyeLmomYanyj

[CANCER RESEARCH54, 3096-3100, June 15, 19941 Advances in Brief Pharmacokinetics of the in Vivo and in Vitro Conversion of 9-Nitro-20(S)- camptothecin to 9-Amino-20(S)-camptothecin in Humans, Dogs, and Mice' Helimuth R. Hinz,2 Nicholas J. Harris, Ethan A. Natelson, and Beppino C. Giovanella The S:ehlin Foundation for Cancer Research at St. Joseph Hospital, Houston, Texas 77003 Abstract We have determined that 9-nitro-20(S)-camptothecin (9NC) converts to 9-amino-20(S)-camptothecin (9AC) in humans, dogs, and mice. Following a single oral dose of 0.1 mg/kg of 9NC, the human plasma concentration reached a maximum concentration of 483 ngJml at 3.4 h with an area under the curve (AUC) of 2.6 @ig-h/ma1nd a half-life of 2.5 h. As conver slon of 9NC to 9AC occurred, the maximum calculated concentration of 9AC was 14.0 ng/ml at 10.3 h with an AUC of3ll ng-h/ml and a half-life of 7.1 h. Following a single oral dose of 1.0 mg/kg of 9NC, the maximum concentration of9NC In the human volunteer was 1247 ng/ml at 53 h with an AUC of 17194 ngh/malnd a half-life of 4.9 h. In this human, the C,,,@ of 9AC was 208 ng/ml at 17.2 h; the AUC was determined to be 9121 ng-h/ml, and the half-life was 13.1 h. In a dog after a single oral dose of 1.0 mg/kg 9NC, the maximum concentration for 9NC was 19.1 ng/ml at 0.7 h with a half-life of 6.4 h and an AUC of 186 ng-h/ml. The maximum concentration of 9AC In this dog was 9.2 ng/ml at 2.9 h with an AUC of 310 ng-h/ml and a half-life of2l.1 h. The maximum concentration of 9NC in the mouse after a single oral dose of4.1 mg/kgof9NC was 732ng/mlat time 0.1 h with an AUCof 441 ag-h/mi and a half-life of 10.0 h. The maximum concentration of 9AC in the mouse was 26 ng/mI at 0.6 h. The AUC was 63 ng-h/ml, and the half-life was 1.2 h. Incubation of mouse liver, spleen, kidney, brain, and muscle tissue with 9NC all indicated conversion to 9AC, yet no conversion was observable in cell-free plasma from human or mouse blood. Structural 9AC was confirmed by mass spectrometry. identification of eliminate much of the severe toxicity which was observed with the water-soluble carboxylate form (22). At the same time, we reported (7) that 9AC and 9NC showed greater antitumor activity than CPT. When establishing the plasma levels of 9NC after oral administration to dogs, we found that 9NC converts to 9AC. In this report, we describe the results of our studies on this conversion in different species in vitro as well as in vivo. Materials and Methods ChemicaLs CPT was imported from Good Land Enterprises, Ltd. (Vancouver, British Columbia, Canada) or purchased from Sigma Biochemicals (St. Louis, MO). Regardless of the source, CPT was purified to 99% for all studies. Acetic acid, acetonitrile, chloroform, methanol, ammonium acetate, sodium azide, N,N dimethylacetamide were ACS quality reagents and obtained from Aldrich Chemical Co. (Milwaukee, WI). Thin layer chromatography plates were from Eastman KOdak Company (Rochester, NY). Absolute ethanol (200 proof) was from AAPER Alcohol Co. (Shelbyville, KY). Perchloric acid was from Fisher Scientific (Pittsburgh, PA). HPLC grade water was prepared in-house using the Mille-Q g.tiPlus system from Millipore(Bedford, MA). HPLCbufferswere ifitered through a 0.45-sm nylon filter from Schleicher and Schuell (Keene, NH) before use. Reduced pentacarbonyl iron was from Sigma Biochemicals. RPM! 1640 tissue culture medium was without fetal calf serum. Introduction Instrumentation CPT3 (1) has been a focus of attention in cancer research for almost three decades, since it first was shown to have considerable antitumor activity against L1210 mouse leukemia (1). Unfortunately, the CPT derivative selected for clinical trials was the sodium carboxylate salt, used because it was water soluble and could be injected via the i.v. route. In this form, the CPT molecule produced considerable toxicity The HPLC system consisted of a Beckman 421 controller with two 11OA pumps and a 2000-@ilinjection loop. The UV and the fluorescence detectors were SPD-11OAV and RF 551, respectively. Both detectors were from Shi madzu (Kyoto, Japan). Reverse phase HPLC analysis of the samples was carried out by using a mobile phase consisting of 23% methanol:acetonitrile (3:7) and 77% of 10 mM of ammonium formate at pH 1.95. Analyses were carried out at room temperature with a flow rate of 1 mI/mm. The HPLC and seemed devoid of anticancer activity (2"6a) s we have also shown detectors were connected in series and set to monitor the UV absorbance at 220 in our laboratory (7). Several CPT analogues have been synthesized nm. The fluorescence detector excitation was set to 360 nm, and the emission over the years (8"17),but interest in camptothecin and its derivatives suddenly increased after it was discovered that camptothecin inhibits topoisomerase I (18"21).Using native CPT with the closed lactone ring intact, we have demonstrated in our laboratory that the oral and intramuscular routes of drug administration are superior to i.v. injec tions. Thus, we were able to maximize the antitumor activity and was monitored at 454 nm. The integrating software used was EZChrome (Shimadzu) and FLO-ONE\Beta (Radiomatic Instruments, Meridian, Cl'). A dual-channel model BD112 flatbed recorder was from Kipp and Zonen (Bo hemia, NY). The samples were usually analyzed on a C@-MicrosorbMV HPLC column or on a cyano or a C18-MicrosorbMV HPLC column (Rainin Instru ments, Woburn, MA). Samples were collected for further analysis with a HeliFrac from Pharmacia (Piscataway, NJ). Samples were weighed on a model Received2/25/94;accepted5/4/94. The costs of publicationof this articlewere defrayedin partby the paymentof page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. H1OT by Mettler Instrumentation Corporation (Highstown, NY). Pharmaco kinetic computations were carried out with RSTRIP II software from Micro math Scientific Software (Salt Lake City, UT). The normal and the tandem mass spectra were taken on a ZAB-SEQ Mass Spectrophotometer from V. G. I This work was supported by funds from The Stehlin Foundation for Cancer Research. 2 To whom requests for reprints should be addressed, at The Stehlin Foundation Cancer Research, 1315 Calhoun Street, Suite 1818, Houston, TX 77003. for 3 The abbreviations used are: CPT, camptothecin; 9AC, 9-aminocamptothecin; 9-nitrocamptothecin; HPLC, high performance liquid chromatography. 9NC, Analytical of Manchester, England at Baylor College of Medicine, Department of Experimental Therapeutics. Dogs were obtained from St. Joseph Surgical TrainingCenter(Houston,TX). Swiss nude mice were bredin our laboratory as described before (23). 3096 9-NITROCAMPTOTHECINTO 9-AMINOCAMPTOTHECINCONVERSION Drug Administration tissueThe Table 1 Distributionsof9NC and 9AC in Drug administration was carried out orally in dogs and humans by packag ing the drug in a gelatin capsule. In the mouse experiment, CPT was given by theoralrouteby suspendingthedrugin cottonseedoil at a concentrationof 4.0 mg/kg/0.1 ml, followed by administrationof 0.1 ml by trochar.Nude mice of Swiss background weighed approximately 30 g. The dog was healthy and mincedraenldativeconversionof 9NC to 9AC in five differenttissues. Tissues were wasafudrdtehder homogenized in a 2-mi Pyrex tissue grinder. To 1 g ofhomogenized tissue Thissol5utmiolnof RPMI solution without fetal calf serum containing 2500 ng of 9NC. bythe was incubatedat 37CA. liquotswere takenout afterperiodof time, extracted estaHblPisLhCed.Tpirsoscueedure,and injectedonto (ng/ml)CBrain Time (.@)a % conversionb 9AC weighed 20 kg. After obtaining permission from the St. Joseph Hospital Medical Research Committee (Institutional Review Board), a healthy human volunteer(B. C. 0.), weighing 90 kg, receiveda single oraldose of 9NC at0.1 mg/kg and 1.0 mg/kg with a three-week interval between each dose. Chemistry 317Liver-fresh 299Liver-boiled 0Spleen 285Muscle 337Kidney 37.8Whole TracebloodPlasma 24 1 0.15 30 24 1 24 63.4 59.8 0 57 67.4 75.6 Trace HPLC Conditions. Using a C@column, the mobile phase consisted of 23% acetomtrile:methanol (7:3) and 77% of 10 mM ammonium formate (pH 1.95) for all determinations.The detection limit under these conditions was 0.25 24 0 0 a Time after which the maximum concentrationswere observed. b Value reached for the maximum percentage of conversion of 9NC to 9AC. C Measured amounts 9AC in ng/ml. No conversion was observed in blood plasma. ng/ml when analyzing a sample for 9AC only. When the amount of total drug was measured on HPLC following the conversion of 9NC to 9AC in plasma, min. The plasma was centrifuged for 2 mm at 13,000 X g. The supernatant was the detection limit dropped to 0.5 ng/ml, probably owing to the needed extra drawn off and added to 1000 pi of 10 msi ammonium formate (pH 1.95). manipulation. Perchloric acid (100 @1l;.5 M)was then added and vortexed. Of this solution, In Vitro Conversion of 9NC to 9AC in Dog Liver. Fresh liver from an 500 @.wd as injected into a 2-ml injection loop onto a Microsorb C8 column. adult dog (555.4 g) was weighed and dissected. Of this mass, 278.4 g was The total drug concentration was again measured by comparing the peak areas removed and combined with 600 ml of RPM! media without fetal calf serum to the previously established 9NC standards. and homogenated in a Singer Turboblend 7 blender for approximately three It should be noted that the position and the plasma concentrations of 9NC miii. A 50-nil fraction was separated and used for the control experiment. The and 9AC were also confirmed by the above experiment using different mobile remaining mixture was transferred into a container and stirred magnetically. A phases. In another experiment, the peak corresponding to 9AC was isolated solution of 25 mg of 9NC in 15 ml of an aqueous ethanol solution (33%) using the C8 column and reinjected onto a C18 column. Injection of a previ containing0.1% acetic acid was added.The mixturewas incubatedat 37C. ously collected 9AC peak onto a cyano column also resulted in retention times Fractiouswere collectedat 0, 0.25, 0.5, 1, 2, 3, 4, and 16 h. Sampleswere and the fluorescence intensities that corresponded to the 9AC standard. Fur analyzed for total drug only without attemptingto separatelactone and car thermore, a plasma peak corresponding to 9AC was isolated using the C8 boxylate forms of either 9NC or 9AC. Therefore, each sample was prepared for column and reinjected under gradient conditions, whereby 10% of the above analysisby HPLCby takinga lOOjil fractionof the homogenateT. o this was aqueous buffer was ramped up to 50% over a period of 40 min. In all cases, the added 10 p1 of 1.5 Mperchloric acid, followed by vortexing the mixture for 20 identity and quantity of 9NC and 9AC was confirmed (data not shown). s, followed by equilibration at room temperature for 10 mm. Next, 300 @olf methanolwere added, and the sample was vortexed for 10 s, centrifugedat Phermacokinetic Analysis 13,000 X g for 2 mm, followed by removal of the supematant. The supematant was added to 1000 @odf 10 mMammonium formate with a pH 1.95. A volume The pharmacokinetic parameters were based on the intensities in the fluo of 50 g.dwas injected into HPLC. The retention time for 9AC was 6.1 min. rescence spectrum. The plot of the plasma concentration versus time was used The same technique was followed for the determinations in mouse organs to calculate the terminal rate constant (k) by means of a logarithmic-linear (Table 1). analysis.All integralswere calculatedto infinity.The least squaresminimi In Vivo Determination of 9AC Plasma Levels. Perchloric acid (10 @.tl) zation techniques were carried out using the Powell algorithm. The half-life was added to 100 p1 of plasma sample. The mixture was vortexed for 10 s and values were calculated using the rate constant (k) in the equation t1,@ left at room temperature for 10 mm. Methanol (300 @dw) as added, and the 0.693/k. The calculated value for the area under the curve was accepted when mixture was vortexed for 10 s, followed by centrifugation at 13,000 X g for 2 an independent calculation of the trapezoidal integral agreed with the value mm. The supernatantwas removed and added to 1000 pAof 10 m@iammonium obtained for the area under the curve, allowing a deviation ofless than 5%. The formate (jH 1.95). Of this solution, 500 ,.d was injected through a 2-ml best curve fitting was obtained for all three vehicles when second order kinetic injection loop onto a Rainin Microsorb C@column. The 9AC concentration was modelswere used. determined by measuring the fluorescence area using an excitation wavelength of 360 urnandan emission wavelengthof 454 am andcomparingthe spectral Mnss Spectrometric Analysis peaks to the previously established standards. These standards were calculated according to the formula: The biological fractions from the dog liver incubation experiment that Fluorescence area "A ng/ml= B corresponded to the peak of 9AC were pooled from HPLC and lyophilized. Storage was carried out under nitrogen at "70uCntil analysis, at which time the samples were reconstituted in methanol. Fig. 1 shows the fragmentation with the constantsofA = 4538.14 andB 835.04. The correlationcoefficient was 0.9977, and the value for R2 was 0.9955. Plasma samples were obtained at different time points for humans, dogs, and mice. Since the overall fluores cence of 9NC is less than the fluorescence of 9AC, the in vivo content of 9NC in the blood plasma was determined by reducing 9NC in situ. This allowed for pattem of the tandem mass spectrum of the biological liver sample with a ru/e 363.@ [C@H17N3O4]''. The top graph of Fig. 2 shows the fragmentation patters of the biological sample, and below it is shown an authentic sample of 9AC as prepared in our laboratory, showing that both samples are indeed identical. the determination of the total drug concentration (9NC plus 9AC). The levels Results of 9NC were then determinedby subtractingthe previouslydetermined9AC levels. Pharmacological Results. In general, our efficiency of converting Procedure for the Determination of Total Drug. To 100 pJ of plasma 9NC to 9AC in either organic or aqueous solution was about 100%. samplewas added20 @o&flanaqueousironsolutionconsistingof 5 ml of water and 100 mg of reduced pentacarbonyl iron, followed by the addition of 20 @&l of concentratedHG. The mixturewas vortexedfor 10 s andallowedto remain 4 J. S. Stehlin, E. A. Natelson, H. R. Hinz, B. C. Giovanella, P. D. De Ipolyi, K. M. Fehir, T. P. Trezona, D. M. Vardeman, N. J. Harris, A. K. Marcee, A. J. Kozielski, and at roomtemperaturefor 10 mm. Then300 @o.df 10%ammoniumhydroxidein A. Ruiz-RazuraP. haseI clinicaltrialandpharmacokineticrsesultswithoraladministra methanol was added, followed by vortexing for 10 s and equilibrating for 5 tion of 20(S)-camptothecin,submittedfor publication. 3097 @ S- -- l@ -s@ .-.-- 9-NITROCAMPTOThECIN TO 9-AMINOCAMFTOThECIN CONVERSION IN N N 75 7' 65 6$ 55 .1 5' 45 255.2 4' 35 I 25 2@ Is @ IS 731 1541 WI 315.1 331.2 363.2 2 @ S, J5@?L1-i'@ J@,@l1l @d@1LT@lkL.L@L.Lld 234@ 281 1 14.2 iii 346 @ __________. 97 1155.1 Il .1 _______ @JL4L@_4_@__________ 5I IN 2N Fig. 1. Shows the detailed mass spectrum of the biological blood plasma sample. The molecular ion appears at 363.2, corresponding to the molecular formula of 9AC. IP U a I a II a i. U - m . as _________________ I a @-. II @@ I I 21 - "@-@---@-- 2U@' 251 -@- - ___ Fig. 2. Top, fragmentation pattern of the plasma sample. Bottom, an authentic 9AC sample was synthesized in our laboratory by reduction of 9NC to 9AC. The tracings reveal the same identity of both samples. 3098 9-NITROCAMPTOTHECINTO 9-AMINOCAMPTOThECIN CONVERSION C Graph 2 @ -.-...-...- Fig. 3. Elimination curves after oral administra tion of 9NC. All four tracings use the Y-log scale to show the amounts of 9NC and 9AC in the blood plasma. The Graph 1 (upper left) shows human plasma after a single dose of 0.1 mg/kg. Curve A shows levels of 9NC, and Curve B shows 9AC concentration. The curve was calculated with a weight factor of 0.8. Graph 2 (upper right) shows human plasma concentration after single dose of 1.0 mg/kg. Curve C shows the levels of 9NC, and Curve D shows 9AC concentration. The curve was calculated with a weight factor of 0.2. In Graph 3 (lower left), the dog plasma concentration is shown after single dose of 1.0 mg/kg. Curve E shows the levels of 9NC, and Curve F shows 9AC concentra tion. The curve was calculated with a weight factor of 0.2. In Graph 4 (lower right), the mouse plasma concentration is shown after administration of a single dose consisting of 4.0 mg/kg. Curve G shows the levels of 9NC, and Curve H shows 9AC con centration. The curve was calculated with a weight factor of 1.0. ng/nIL B o@",.-"--- p., , . I-"@mL : @II@..TI@HII@11' IF. 1.lu@ 13@' 0.00 n.oo 20.00 30.00 40.00 50.c Time (hoors) 0.00 8.00 16.00 Tr(uh@zs) @. 24.00 0.00 10.00 20.00 Time (hours) 30.tX 0.00 8.00 16.00 Time (hours) 24.00 Our experimental efficiency of monitoring the 9NC conversion by of the curves was seen in dogs (Fig. 3, curves E and F), but here no first reducing the plasma sample in situ, followed by the extraction, toxicity was observed. was approximately 90%. The results of the pharmacokinetic analysis Our results from the incubation experiment are summarized in of the blood plasma samples are summarized in Table 2 and graphi Table 1. From this experiment, we suspected that this reaction could cally displayed in Fig. 3. When the three systems of humans, dogs, be enzymatic in nature and probably due to the action of a reductase. and mice are compared, it becomes obvious that their metabolism of This conclusion is supported by the observation that boiling the above 9AC and 9NC are quite different. This is most obvious when com liver homogenates for 10 mm eliminated the ability to cause the paring the elimination curves of 9AC and 9NC to each other, wherein conversion of 9NC. Further, incubation of 9NC with fresh human it was found that the mouse has the ability to eliminate 9NC and 9AC plasma yielded no conversion to 9AC, even after a 48 h incubation the quickest. When comparing the curves and the calculated pharma period. All of the other tested tissues (Table 1) showed appreciable cokinetic values of the human high dose (1.0 mg/kg) with its low dose conversion of 9NC to 9AC, complicating the overall pharmacokinetic (0.1 mg/kg), it becomes evident that: (a) the blood does get saturated picture. with 9NC only at higher dose levels; and (b) the drug stays longer in Toxicity. The human toxicity observed at the higher dose of 1.0 the blood. This can be seen from Fig. 3, curves C and D, where both mg/kg was total alopecia with a drop in the total leukocyte count from curves meet at approximately 24 h. We suspect that this slow elimi 5,300/mm3 to 3,000/mm3, and a fall in the platelet count from nation after saturation with 9NC and 9AC at 1.0 mg/kg was respon 230,000/mm3 to 70,000/mm3. Hematological recovery occurred sible for the observed toxicity in the human model. The same crossing within a week. The lower dose of 0.1 mg/kg of 9NC did not affect the blood count. No animal toxicity was observed. This problem of toxicity should be clinically manageable, however, since we have mousePharTmaabcleok2inPehtiacrmacokinetic results for human, dog, and datapoints results for human, dog, and mouse. The best curves for the model were obtained when the data wceurevef.iCt mfo,r,,a, biexponential observed good antitumor activity at much lower levels5 in animal models. A major problem with 9AC is that its chemical synthesis, using the (hIonurvsiv)Ho usmysatnem-9NC (time in hours)b AUC' t1,@ semisynthetic method, is carried out by nitration of CPT, followed by reduction to the amino group. This is a low yield operation. In 2.5Human-9A(C0.1) 7.1Human-9N(0C.1) 4.9Human-9A(C1.0) 13.1Dog-9NC(1.0) 6.4Dog-9AC(1.0) 21.1Mous(e1-.09)NC 10.0Mouse-(94A.1C) (4.1) 483 (3.4) 14.0(10.3) 1,247 (5.3) 208 (17.2) 19.1 (0.7) 9.2(2.9) 732(0.1) 26 (0.6) 2,599 311 17,194 9,121 186 310 441 63 addition, 9AC is light sensitive, heat sensitive, and oxygen sensitive. This makes the production and stabilization of 9AC difficult. The decomposition reactions of 9AC lead to compounds that exhibit a large degree of toxicity in nude mice, whereas the toxicity of pure 9AC is significantly less.5 We were, therefore, interested in the use of 1.2 9NC as a clinical alternative to 9AC, because of its easier availability a Type of test system used. The numerical values in parentheses give the magnitude of the single dose in mg/kg. and greater stability. b Area under the curve in ngh/ml. C Maximum plasma concentration in ng/ml. 5 Unpublished results. 3099 9-NITROCAMFFOThECIN TO 9-AMINOCAMPTOTHECINCONVERSION Discussion We believethat 9NC would make an excellentchoicefor use in a Phase I clinical trial for several reasons: (a) its antitumor activity is greater by a factor of 3 than the antitumor activity of CPT in our nude mouse model; (b) the pharmacological effects in the human of a single dose (1.0 mg/kg) show a prolonged therapeutic interval available with this drug. 9NC has shown anticancer activity at plasma concentrations as low as 1 ng/ml maintained for 24 to 96 h, using our human xenograft model. We have observed that we can maintain 100 times this concentration for 48 h in a human subject with a peak concen tration equaling over 1000 times the therapeutic dose and the obser vation of only minimal toxicity. For instance, we did not see any signs of cystitis or diarrhea as we saw during our clinical study with CPT. We would expect to use lower amounts of 9NC in order to obtain good antitumor effects. We are now investigating whether the anti cancer activity seen after 9NC administration is due to 9NC itself or by the 9AC generated by the conversion reaction described above. Acknowledgments We thankDr. Kevin Ballardfor takingthe mass spectra. References 1. Wall, M. E., Wani, M. C., Cook, C. E., Palmer, K. H., McPhail, A. T., and Sim, G. A. Plant antitumor agents. I. The isolation and structure of camptothecin, a novel alkaloidal leukemia and tumor inhibitor from Camptotheca acuminata. J. Am. Chem. Soc., 88: 3888"38901, 966. 2. Gottlieb, J. A., Guarino, A. M., Call, J. B., Oliverio, V. T., and Block, J. B. Preliminary pharmacologic and clinical evaluation of camptothecin sodium (NSC 100880). Cancer Chemother. Rep. Part 1, 54: 461"4701, 970. 3. Gottlieb, J. A., and Luce, J. K. Treatment of malignant melanoma with camptothecin (NSC-100880). Cancer Chemother. Rep., 56: 103"1051,972. 4. Muggia, F. M., Creaven, P. J., Hansen, H. H., Cohen, M. H., and Selawry, 0. S. Phase I clinical trial of weekly and daily treatmentwith camptothecin(NSC-100880): correlation with preclinical studies. Cancer Chemother. Rep., 56: 515-521, 1972. 5. Moertel, C. G., Schutt, A. J., Reitemeier, R. J., and Hahn, R. G. Phase II study of camptothecin (NSC-10088O) in the treatment of advanced gastrointestinal cancer. Cancer Chemother. Rep., 56: 95"1011,972. 6. Schaeppi, U., Fleischman, R. W., and Cooney, D. A. Toxicity of camptothecin (NSC-100880). CancerChemother.Rep. Part3, 5: 25"361,974. 7. Giovanella, B. C., Stehlin, J. S., Wall, M. E., Wani, M. C., Nicholas, A. W., Liu, L F., Silber, R., and Potmesil, M. DNA topoisomerase-Itargetedchemotherapyof human colon cancer in xenografts. Science (Washington DC), 246: 1046"10481,989. 8. Plattner, J. J., Gless, R. D., Cooper, G. K., and Rapoport, H. Synthesis of some DE and CDE ring analogs of camptothecin.J. Org. Chem.,39: 303"3111,974. 9. Danishefsky,S., andEtheredge,S. J. Synthesisandbiological evaluationof DE-AB camptothecin. J. Org. Chem., 39: 3430"34321, 974. 10. Bristol, J. A., Comins, D. L, Davenport, R. ~WK.,ane, M. J., Lyle, R. E., Maloney, J. R., and Portlock,D. E. Analogs of camptothecin.J. Med. Chem., 18: 535-537, 1975. 11. Sugasawa, T., Toyoda, T., Uchida, N., and Yamaguchi, K. Experimentson the synthesis of dl-camptothecin. 4. Synthesis and antileukemic activity of dl-camptoth ecin analogs. J. Med. Chem., 19: 675"6791, 976. 12. Wani, M. C., Ronman, P. E., Lindley, J. T., and Wall, M. E. Plant antitumor agents. 18. Synthesis and biological activity of camptothecinanalogs. J. Med. Chem., 23: 554"5601, 980. 13. Wani,M. C., Nicholas, A. W., andWall, M. E. Plantantitumoragents.23. Synthesis and antileukemic activity of camptothecin analogs. J. Med. (1cm., 29: 2358"2363, 1986. 14. Wani, M. C., Nicholas, A. W., Manikumar, G., and Wall, M. E. Plant antitumor agents. 25. Total synthesis and antileukemic activity of ring A substituted campto thecin analogues. Structure-activity correlations. J. Med. Chem., 30: 1774"1779, 1987. 15. Nicholas,A. W., Wani, M. C., Manikumar,0., Wall, M. E., Kohn, K. W. and Pominier, Y. Plant antitumor agents. 29. Synthesis and biological activity of ring D and ring E modified analogues of camptothecin. J. Med. Chem., 33: 972"9781,990. 16. Kingsbury,W. D., Boehm,J. C., Jakas,D. R., Holden,K. 0., Hecht,S. M., Gallagher, 0., Caranfa, M. J., MCCabe, F. L, Faucette, L F., Johnson, R. K., and Hertzberg, R. P. Synthesis of water-soluble(aminoalkyl)camptothecinanalogues:inhibitionof topoisomerase I and antitumor activity. J. Med. Chem., 34: 98"1071, 991. 17. Ejima, A., Terasawa, H., Sugimori, M., Ohsuki, S., Matsumoto, K., Kawato, Y., Yasuoka, M., and Tagawa, H. Antitumoragents. V. Synthesis and antileukemic activity of E-ring-modified (RS)-camptothecin analogues. Chem. Pharm. Bull. (Tokyo), 40: 683"6881, 992. 18. Hsiang,Y-H.,HertzbergR, .,Hecht,S.,andLiu,L F. Camptotheciinnducesprotein linked DNA breaks via mammalian DNA topoisomerase 1. J. Biol. Chem., 260: 14873"14878, 1985. 19. Hsiang, Y-H., and Liu, L F. Identificationand mammalianDNA topoisomerase I as an intracellulartarget of the anticancerdrug camptothecin.Cancer Res., 48: 1722"172169,88. 20. Jaxel,C.,Kohn,K.W.,Wani,M.C.,Wall,M.E.,andPommierY, .Structure-activity study of the actions of camptothecin derivatives on mammalian topoisomerase. I. Evidence for a specific receptor site and a relation to antitumor activity. Cancer Res., 49: 1465"1469,1989. 21. Hsiang, Y-H., Liu, L F., Wall, M. E., Wani, M. C., Nicholas, A. W., Manikumar, G., Kirschenbaum, S., Silber, R., and Potmesil, M. DNA topoisomerase I-mediated DNA cleavage and cytotoxicity of camptothecin analogues. Cancer Res., 49: 4385"4389, 1989. 22. Giovanella, B. C., Bins, H. R., Kozielski, A. J., Stehlin, J. S., Silber, R., and Potmesil, M. Completegrowthinhibitionof humancancerxenograftsin nudemice by treatment with 20-(S)-camptothecin. Cancer Res., 51: 1779"18031,991. 23. Giovanella, B. C., and Stehlin, J. S. Heterotransplantation of human malignant tumors in oenudethym usless mice. I. Breeding and maintainance of oenudemice. J. Natl. Cancer Inst., 51: 615"6191, 973. 3100