Mitomycin as a Modulator of Irinotecan Anticancer Activity

ABSTRACT: Irinotecan and mitomycin (Mutamycin) possess significant single-agent activity against several tumor types, and mitomycin activates topoisomerase I, the cellular target of irinotecan. We conducted a phase I dose-escalation study of irinotecan and mitomycin in 37 evaluable patients with solid tumors. Antitumor responses included 2 complete responses, 5 partial responses, 10 minor responses, and a CA 19-9 tumor marker response. Responders included 14 patients previously treated with chemotherapy for metastatic disease. No pharmacokinetic interaction between mitomycin and irinotecan was apparent when these agents were given 24 hours apart. Responders (complete and partial responses) demonstrated the largest topoisomerase I induction 24 hours following mitomycin infusion. In addition, since maximum topoisomerase I up-regulation was reached 24 hours after administration of mitomycin, a delay in the administration of irinotecan after mitomycin appeared justified. Based on these encouraging phase I data, phase II clinical trials in breast and esophageal/gastroesophageal junction adenocarcinomas at the recommended doses and schedule are under way. [ONCOLOGY 16(Suppl 7):21-25, 2002]

Irinotecan(CPT-11, Camptosar) is a semisynthetic analog of camptothecin, a compoundoriginally isolated from the Chinese/Tibetan ornamental tree Camptothecaacuminata. In vivo, cellular carboxylesterases cleave the ester bond ofirinotecan, which is inactive, thereby producing the active compound7-ethyl-10-hydroxycamptothecin (SN-38). Irinotecan/SN-38 interacts with cellulartopoisomerase I/DNA complexes and has S-phase-specific cytotoxicity.[1]Irinotecan binds to topoisomerase I single-strand DNA breaks, and thisreversible topoisomerase I/irinotecan/DNA cleavable complex, though not initself lethal to the cells, collides with the advancing replication forks,leading to the formation of a double-strand DNA break, irreversible arrest ofthe replication fork, and cell death.[1] The collision of the irinotecan-topoisomeraseI complex with the replication fork also results in G2 arrest/delay by signalingthe presence of DNA damage to an S-phase checkpoint mechanism.[2]

Because topoisomerase I is the cellular target of irinotecan, itis conceivable that the cellular level of topoisomerase I would be proportionalto irinotecan cytotoxic effects. This notion is supported by experimentalevidence from yeast systems and mammalian cell lines.[3,4] The total activity oftopoisomerase I was reduced (by one-fourth) in irinotecan cell lines renderedresistant by stepwise, continuous treatment with the drug compared to theirinotecan-sensitive parental cell line.[4] Theoretically, the topoisomerase Iexpression in tumor specimens may serve as a predictor for sensitivity toirinotecan chemotherapy. Since tumor cells may escape irinotecan cytotoxiceffects by down-regulating their levels of topoisomerase I, strategies toincrease topoisomerase I expression might enhance the antitumor effect of thisagent.

Mitomycinand Increased Topoisomerase I Activity in Preclinical Systems

Mitomycin (Mutamycin) is an antitumor antibiotic isolated from Streptomyces caespitosus. This agent is activated in vivo to an alkylatingmoiety that covalently crosslinks complementary DNA strands, resulting in theinhibition of DNA synthesis.[5] Metabolic activation of mitomycin isaccomplished by reduction of the quinone moiety by the enzyme DT-diaphorase(NQO1), which releases a methanol residue from the molecule.

Pilot data from our laboratory demonstrated that irinotecandecreases NQO1 in vivo gene expression in peripheral blood lymphocytes byapproximately 50%, suggesting that pretreatment with irinotecan may interferewith mitomycin activation.[6] These findings are consistent with the observedlack of potentiation of mitomycin activity by pretreatment (24 hours prior) withirinotecan in the EH-6, H-111, and CH-6 human tumor xenografts.[7]

Gobert et al studied topoisomerase I activity in MCF cells underconditions in which p53 expression was induced by mitomycin.[5] Theseinvestigators observed that mitomycin increased topoisomerase I activity asmeasured by relaxation of supercoiled DNA and by phosphorylation of SR proteinsplicing factor. This increase in catalytic activity occurred in conjunctionwith the nuclear accumulation of p53, resulting in detectable activation oftopoisomerase I within less than 1 hour of drug treatment. Further studiesdemonstrated that the interaction between p53 and topoisomerase I is observedwith both latent and activated wild-type p53, as well as with several mutant andtruncated p53 proteins in vitro.[8]

Because both irinotecan and mitomycin possess significantsingle-agent activity against several tumor types, and mechanistically mitomycinmay result in activation of topoisomerase I, we evaluated the combination ofirinotecan and mitomycin in a phase I clinical trial. We hypothesized thatadministering mitomycin prior to irinotecan would permit mitomycin activationwithout interference and might result in increased topoisomerase Iexpression/activity, leading to increases in tumor cell sensitivity toirinotecan. Because of possible activation interference, irinotecan wasadministered 24 hours after mitomycin.

Patients and Methods

Patients with histologically confirmed advanced solidmalignancies were candidates for this study. Eligibility criteria also includedadequate hematopoietic (absolute neutrophil count [ANC] ³ 1,500/µL, plateletcount ³ 100,000/µL, and hemoglobin ³9.0 g/dL), hepatic (total bilirubin level< 1.5 mg/dL; transaminases [AST, ALT] and alkaline phosphatase £3 times the upper limit of normal), and renal (serum creatinine £1.5 mg/dL) functions. No prior treatment with mitomycin, irinotecan, ornitrosourea, and no more than six courses of chemotherapy containing analkylating agent (four courses for carboplatin [Paraplatin]) were permitted, aswell as no prior irradiation to more than 20% of bone marrow reserve. Due topossible interference with activation of mitomycin, concomitant treatment withcoumarin anticoagulants was not permitted. All patients gave informed writtenconsent.

Treatment Plan andDose Escalation

The dose schedule was mitomycin on day 1 and irinotecan on days2 (24 hours after mitomycin), 8, 15 and 22, with cycles repeated every 6 weeks(actual recovery period of 21 days). Some patients experienced late diarrhea,which caused delays in late doses of irinotecan. As a result, the scheduleduration was decreased to 4 weeks, with irinotecan administered on days 2 and 8after mitomycin on day 1. The starting dose of irinotecan was 50 mg/m²administered IV over 90 minutes, whereas the dose of mitomycin was kept at 6mg/m². The number of cycles of mitomycin was limited to a maximum of six (36mg/m² total dose) to avoid potential cumulative toxicities of mitomycin.

Dose escalation of irinotecan was to proceed in 25-mg/m²increments in cohorts of at least three new patients until dose-limitingtoxicity was observed in at least two patients. Dose reduction by one level wasallowed for patients who experienced dose-limiting toxicity or grade 3/4diarrhea. The recommended dose (maximum tolerated dose) was defined as thehighest dose at which no more than one of six new patients developeddose-limiting toxicities during the first course. The dose-limiting toxicity wasdefined as: (1) ANC < 500/µL lasting at least 5 days, or associated withfever; (2) grade 4 thrombocytopenia; (3) nonhematologic toxicity  ³grade 3,except diarrhea or nausea/vomiting; and (4) grade 4 diarrhea despite optimalantidiarrheals, or grade 4 vomiting despite optimal antiemetics.

Pharmacokineticand Molecular Correlates Sampling and Analysis

Blood was sampled from a site contralateral to the drug infusionduring the first course of treatment at 2, 4, and 24 hours postinfusion toevaluate plasma concentrations of irinotecan, SN-38, and SN-38G. TopoisomeraseI, carboxylesterase 1, carboxylesterase 2 (CE1 and CE2), and NQO1 geneexpression were analyzed in peripheral blood mononuclear cells. RNA wasextracted from blood samples collected at the following times: (1) baseline, (2)5 minutes into the mitomycin infusion, (3) at the end of mitomycin infusion, (4)3 hours after the end of mitomycin infusion, (5) 24 hours after end ofmitomycin, (6) at the end of irinotecan infusion, (7) 2 hours after the end ofirinotecan, and (8) 24 hours after the end of irinotecan infusion. Geneexpression samples were analyzed by reverse-transcription polymerase chainreaction with a reaction-specific internal standard. Detection was done bycapillary electrophoresis with laser-induced fluorescence.

Results

Dose Escalationand Toxicities

A total of 38 patients (37 evaluable) from two institutions(Cancer Therapy and Research Center, San Antonio, Texas, and The Arthur G. JamesCancer Hospital and Richard J. Solove Research Institute at the Ohio StateUniversity, Columbus, Ohio) were treated with 119 courses of theirinotecan/mitomycin combination (pertinent demographic characteristics aredisplayed in Table 1). Five patients received 14 additional courses ofirinotecan as a single agent after the maximum allowed number of courses withmitomycin (six) was exceeded. Thirty-two patients had received priorchemotherapy, including 21 with previous radiation treatment (patients withprevious pelvic irradiation were ineligible).

Table 2 depicts the dose escalations and dose-limitingtoxicities in the study. Initially, patients were treated in the 6-week schedule(mitomycin on day 1, irinotecan on days 2, 8, 15, and 22). Although no severetoxicities were observed among seven patients receiving irinotecan at 50mg/m²/wk, two of six patients completing a full course (4 weeks) of irinotecanat 75 mg/m²/wk in this schedule developed grade 3/4 diarrhea. Since the diarrheaoccurred after patients received the third or fourth dose of irinotecan, thestudy design was amended to shorten the courses to 4 weeks (mitomycin on day 1,irinotecan on days 2 and 8).

Higher weekly doses of irinotecan were tolerated in the amendedschedule, with only 1 of 14 patients treated at irinotecan doses of 100 to 125mg/m²/wk developing moderate to severe toxicity (grade 3 diarrhea) during thefirst course of treatment. However, dose-limiting toxicities (febrileneutropenia and grade 3/4 diarrhea) were observed in two of three patientstreated with irinotecan at 150 mg/m²/wk. Therefore, the 125-mg/m²/wk dose levelon the 4-week schedule is the dose and schedule of the combination recommendedfor phase II studies. Other mild to moderate toxicities includednausea/vomiting, asthenia/fatigue, and peripheral edema. No renal dysfunction,hemolysis, or pulmonary interstitial fibrosis occurred. Two patients died onstudy, one from congestive heart failure and another from pulmonary embolism(necropsy proven).

Antitumor Activity

Of 37 evaluable patients, 7 had major responses. This includedtwo complete responses and five partial responses. An additional 10 patients hada minor response. Among patients with any degree of response, 14 had previouschemotherapy for metastatic cancer. The types of tumor in which activity wasdetected included refractory breast cancer (one complete response, one partial,one minor), chemoradiation-resistant esophageal cancer (one complete response [Figure1], two partial, two minor), gastric carcinoma (one partial response,two minor), previously treated non-small-cell lung cancer (one partialresponse, two minor, and prolonged disease stabilization in one patient),pretreated pancreatic carcinoma (one minor response, one tumor marker response),cholangiocarcinoma (one minor response), and hepatocellular carcinoma (one minorresponse).

Pharmacokinetics

Pharmacokinetic evaluation was performed by Dr. L. Schaaf(Pharmacia Corporation, Peapack, New Jersey) and Dr. J. Kuhn (University ofTexas Health Sciences Center, San Antonio, Texas). Irinotecan and SN-38 mean t_1/2 were 5.7 ± 0.7 and 10.7 ± 4.4 hours, respectively, and clearance was 12.7± 3.86 L/h/m² for irinotecan. The SN-38/irinotecan AUC_0-24 ratio was 0.03 ±0.01 (n = 29). Irinotecan, SN-38, and SN-38G pharmacokinetic parameters weremore similar at 100 and 125 mg/m² than at historical controls, which indicatedthat no pharmacokinetic interactions occurred between mitomycin and irinotecanon the schedule tested in this study.

Molecular Correlates

Topoisomerase I gene expression at baseline was 516 ng/µL ± 83with maximal induction (1,863 ng/µL ± 310) by 24 hours after mitomycininfusion. Patients with major responses to therapy (complete and partialresponses) had the greatest topoisomerase I induction, to 6,114 ng/µL ± 1,019compared to the rest of the patients at 151 ng/µL ± 25.2 (P = .023). Mitomycininduced NQO1 gene expression from a baseline level of 4.0 ng/mL ± 0.67 tomaximum induction of 18 ng/µL ± 3.12 by 3 hours after mitomycin infusion.Irinotecan induced the expression of CE1 from a baseline level of 0.23 ng/µL ±0.04 to a maximum induction of 1.90 ng/µL ± 0.30 by 3 hours after irinotecaninfusion, whereas neither irinotecan nor mitomycin induced CE2.

Discussion

Our clinical data suggest that the mitomycin/irinotecancombination is feasible and active in patients with refractory malignancies. Nopharmacokinetic interactions between mitomycin and irinotecan are apparent whenthese agents are given 24 hours apart. As predicted from in vitro models,mitomycin induces topoisomerase I gene expression in human subjects, andresponders (complete and partial responses) demonstrate the largesttopoisomerase I induction 24 hours following mitomycin infusion. Mitomycin alsoinduces NQO1, while irinotecan decreases NQO1 gene expression shortly afteradministration, which suggests that a serum irinotecan-free interval shouldfollow mitomycin treatment.[9,10] In addition, since maximum topoisomerase Iup-regulation is reached 24 hours after administration of mitomycin, a delay inthe administration of irinotecan after mitomycin appears justified.

Due to the encouraging antitumor activity observed in this phaseI study, phase II clinical trials in breast and esophageal/gastroesophagealjunction adenocarcinomas at the doses and schedule recommended in this study areunder way.

References:

1. Liu LF, Desai SD, Li TK, et al: Mechanism of action ofcamptothecin. Ann NY Acad Sci 922:1-10, 2000.

2. Shao RG, Cao CX, Zhang H, et al: Replication-mediated DNAdamage by camptothecin induces phosphorylation of RPA by DNA-dependent proteinkinase and dissociates RPA:DNA-PK complexes. EMBO J 18:1397-1406, 1999.

3. Reid RJ, Benedetti P, Bjornsti MA: Yeast as a model organismfor studying the actions of DNA topoisomerase-targeted drugs. Biochim BiophysActa 1400:289-300, 1998.

4. Kanzawa F, Sugimoto Y, Minato K, et al: Establishment of acamptothecin analogue (CPT-11) -resistant cell line of human non-small cell lungcancer: Characterization and mechanism of resistance. Cancer Res 50:5919-5924,1990.

5. Gobert C, Bracco L, Rossi F, et al: Modulation of DNAtopoisomerase I activity by p53. Biochemistry 35:5778-5786, 1996.

6. Kolesar J, Villalona-Calero M, Eckhardt G, et al: Detectionof a point mutation and the alternative splice NQO1 gene in patients with coloncancer (abstract 402). Proc Am Soc Clin Oncol 15:189a, 1996.

7. Kim R, Hirabayashi N, Nishiyama M, et al: Experimentalstudies on biochemical modulation targeting topoisomerase I and II in humantumor xenografts in nude mice. Int J Cancer 50:760-766, 1992.

8. Gobert C, Skladanowski A, Larsen A: The interaction betweenp53 and DNA topoisomerase I is regulated differently in cells with wild-type andmutant p53. Proc Natl Acad Sci USA 96:10355-10360, 1999.

9. Traver RD, Horiskoshi T, Danenburg K, et al: NAD(P)H: Quinoneoxidoreductase gene expression in human colon carcinoma cells: Characterizationof a mutation which modulates DT-diaphorase activity and mitomycin Csensitivity. Cancer Res 52:797-802, 1992.

10. Kolesar JM, Burris HA, Kuhn JG: Detection of a pointmutation in NQO1 (DT-diaphorase) in a patient with colon cancer. J Natl CancerInst 87:1022-1024, 1995.