Rational Design of Irinotecan Administration Based on Preclinical Models

Introduction

The antitumor activity of the plant alkaloid camptothecin has been known for more than 2 decades, but initial enthusiasm for this compound was tempered by the recognition of its severe side effects. (Hematologic toxicities are dose-limiting.)[1-5] The discovery that camptothecin perturbs the catalytic cycle of the topoisomerase I enzyme offered a novel target for chemotherapy and revitalized interest in this drug.[6-9]

Topoisomerase enzymes help regulate translation and transcription processes by controlling the topologic structure of DNA and relaxing the supercoiled DNA helix. The topoisomerase I catalytic cycle involves transient covalent binding of the enzyme to DNA (cleavable complex formation), which nicks one DNA strand. This, in turn, allows for DNA strand passage, religation of the DNA strand break, and subsequent release of the enzyme from the DNA. Consequently, the DNA helix is unwound, relieving torsional stress associated with, for instance, DNA replication. Camptothecin and most of its analogs act by mediating stabilization of the topoisomerase I/DNA cleavable complex,[8,9] initiating a series of events that ultimately lead to cell death.

The renewed interest in camptothecin resulted in the development of camptothecin derivatives, which display less severe side effects[10] and better water solubility than the parent compound. Topotecan(Drug information on topotecan) (Hycamtin) and irinotecan(Drug information on irinotecan) (CPT-11 [Camptosar]) are two of the most prominent topoisomerase I-interactive agents for which antitumor activity has been evaluated in clinical trials.

The design of clinical drug regimens of irinotecan has been based on classic in vivo pharmacokinetic and pharmacodynamic considerations, which call for the administration of maximum tolerated doses (MTDs) for any given schedule. Preclinical in vivo data of the antitumor activity of irinotecan against human neuroblastoma xenografts, however, have indicated that antitumor activity depends more on the schedule of administration than on drug dose.[11]

Notwithstanding their empirical design, early phase II-III clinical trials evaluating the topoisomerase I-interactive agents topotecan and irinotecan have had encouraging results in patients with heavily pretreated secondary and refractory leukemias, showing indications of response in these otherwise poor responders.[12-16] Moreover, currently ongoing clinical trials in colorectal cancer combining irinotecan with fluorouracil(Drug information on fluorouracil) (5-FU)/leucovorin have shown promising response rates of up to 60%.[17-19]

Many parameters relevant for studying the action of topoisomerase I-interactive agents have been identified. These include: catalytic activity, protein levels, and mutations of the topoisomerase I enzyme; topoisomerase I messenger RNA (mRNA) levels; and DNA-topoisomerase I protein cross-links.

Known resistance mechanisms against topoisomerase I-interactive agents include decreased topoisomerase I protein levels[20] due to posttranslational or posttranscriptional changes or gene rearrangements[21] and topoisomerase I mutations[22,23] affecting cleavable complex formation (decreased catalytic activity) or binding of topoisomerase I-interactive agent.[23,24] (Catalytic activity is preserved, but the topoisomerase I agent-induced stabilization of cleavable complex formation is reduced.) One mechanism specifically related to irinotecan resistance is a change in the activity of carboxylesterase,[25] the enzyme required for the metabolic conversion of irinotecan to its active metabolite, SN-38. Although many of the above resistance mechanisms have been described in different cell line models, their prognostic value for in vivo sensitivity to topoisomerase I-interactive agents has yet to be established.

The demonstration that (temporary) inhibition of DNA/RNA synthesis could protect from topoisomerase I agent-induced toxicity but not from drug-induced formation of protein/DNA cross-links has led to the "DNA-replication fork collision" model. This model assumes that the protein-DNA cross-links are prelethal lesions, which become lethal only when they encounter a moving DNA replication fork. This encounter results in irreparable double-strand DNA breaks[26-28] and chromosomal aberrations.[29-33]

Supporting evidence for this model comes from drug interaction studies of topoisomerase I- and topoisomerase II-interactive agents.[34,35] These studies have shown that simultaneous exposure of cells to camptothecin and etoposide(Drug information on etoposide) results in antagonistic effects, which can be changed to additive effects if the drug exposures are separated by an interval that exceeds the time necessary to restore drug-induced inhibition of DNA and RNA synthesis.[35]

These data suggest that, to a certain extent, inhibition of DNA/RNA synthesis may provide a cell with an opportunity to restore the DNA/protein cross-link lesions before a fatal collision with DNA replication forks occurs. The fact that topoisomerase I- and topoisomerase II-interactive agents are effective as single agents despite concomitant drug-induced inhibition of DNA/RNA synthesis appears to be contradictory. We propose, therefore, that the extent and duration of DNA/RNA-synthesis must be taken into consideration.

Evidence that these parameters may play a role in ultimate drug-induced cytotoxicity may come from another study exploring the interaction between topoisomerase I- and topoisomerase II-interactive agents, which showed that simultaneous drug exposure results in drug synergism[36] rather than antagonism. In this study, very low concentrations of the drugs were used (< IC₁₀, or the concentration inhibiting 10% of cell growth). One theory that may explain the two seemingly contradictory findings suggests that, at the low drug concentrations used in the latter study, the inhibition of DNA/RNA synthesis is insufficient to interfere with the moving replication forks and to affect drug efficacy.

This article describes a series of experiments designed to determine how exposure to topoisomerase I inhibitors (specifically, camptothecin and irinotecan) affect DNA synthesis, and how the extent and duration of this inhibition may be used to optimize scheduling of these drugs. In addition, a preclinical in vivo model is presented in which a 100% complete response (CR) rate can be achieved with irinotecan in a single-drug regimen. This model should prove to be useful to test the hypotheses developed from in vitro data in an in vivo setting.

Materials and Methods

Cell Lines and Xenografts

The human myeloid leukemia cell line HL60 was propagated in RPMI1640 media supplemented with 10% heat-inactivated fetal bovine serum. Cell cultures were maintained at exponential growth by maintaining cell densities below 1 × 10⁶ cells/mL.

Xenografts were initially established by subcutaneous (SC) injection of 200 mL of a cell suspension (1× 10⁷ cells/mL) in 8- to 12-week-old, female athymic nude mice (Sprague Dawley Inc., Indianapolis, Indiana). Subsequently, xenografts were maintained for several generations by SC transplantation of 50-mg nonnecrotic solid tumor tissue.

Drugs

Camptothecin and SN-38 for the in vitro studies were donated by BioNumerik Pharmaceuticals, Inc. (San Antonio, Texas). Stock solutions of each drug (5 mmol) were made in 100% dimethyl sulfoxide (DMSO) and stored at -20 °C until use. Drug dilutions were made from these frozen stocks with media to achieve the desired final concentrations. Irinotecan for the in vivo studies was donated by the Pharmacia & Upjohn Company (Kalamazoo, Michigan) as a sterile solution of 20 mg/mL in 0.9% saline.