The ability of a panel of camptothecin derivatives to access the tumor compartment was evaluated to determine the mechanisms by which the architecture of solid tumors may act to limit their activity. Microregional localization and activity of members of the camptothecin class of topoisomerase I targeting agents, including topotecan, irinotecan, and irinophore C, a lipid-based nanoparticulate formulation of irinotecan, were evaluated over time in HCT116 and HT29 colorectal tumor xenografts. Using native drug fluorescence, their distributions in tissue cryosections were related to the underlying tumor vasculature, tumor cell proliferation, and apoptosis. Topotecan exhibited a relatively uniform tumor distribution; in tissue 100 μm away from vessels, it reached 94% ± 5% of levels seen around blood vessels, whereas irinotecan and irinophore C were found to reach only 41% ± 10% and 5% ± 2%, respectively. Surprisingly, all three agents were able to initially inhibit proliferation uniformly throughout the tumors, and it was their rate of washout (topotecan > irinotecan > irinophore C) that correlated with activity. To explain this discrepancy, we looked at SN38, the active metabolite of irinotecan, and found it to penetrate tissue similarly to topotecan. Hence, the poor access to the tumor compartment of irinotecan and irinophore C could be offset by their systemic conversion to SN38. It was concluded that all three agents were effective at reaching tumor cells, and that despite the poor access to the extravascular compartment of irinophore C, its extended plasma exposure and systemic conversion to the diffusible metabolite SN38 enabled it to effectively target solid tumors. Mol Cancer Ther; 13(11); 2727–37. ©2014 AACR.
The natural compound camptothecin and its synthetic analogues represent a chemically diverse class of biologically active topoisomerase targeting agents with differing physicochemical properties and clinical applications (1–4). To date, two clinically approved analogues exist, topotecan and irinotecan, while there are currently efforts being made to develop a new generation of compounds that exhibit significantly longer systemic exposures (5–8). To understand the mechanisms limiting the biologic activity of these agents, and to test the fundamentals driving the development of the new generation of compounds, we have applied a tumor mapping technique to evaluate both the microregional activity and localization of a panel of camptothecin analogues. The impact of microregional drug distribution within solid tumors on the efficacy of anticancer drugs is often poorly understood, especially with respect to whether anticancer drugs reach all the cells within a tumor at concentrations high and long enough to exert a therapeutic effect. Factors affecting the delivery and distribution of drugs in tumors span a diverse set of mechanisms, including drug pharmacokinetics as well as tumor vessel functionality, access the tumor cell compartment, uptake, and metabolism by cells (9–12).
We selected for evaluation topotecan, irinotecan, and irinophore C, each having different physicochemical properties of whose impact on tumor microregional localization is largely undefined. In the United States, topotecan is used for metastatic ovarian, cervical, and small cell lung cancer, whereas irinotecan is used for metastatic colorectal cancer. Irinotecan is generally given once per week followed by 48 hours of fluorouracil under the FOLFIRI treatment regimen, whereas topotecan is typically administered daily for 3 to 5 days (13, 14). Preclinical single-agent studies have indicated that both irinotecan and topotecan work best when given repeatedly several days in a row (15–17). Irinophore C, a lipid-based nanoparticulate formulation of irinotecan with excellent preclinical efficacy, was included as being a representative of the next generation of high-molecular weight camptothecin derivatives and nanoformulations. In the case of irinophore C, acidified liposomes in the range of 40 nm are used as a drug carrier platform that produces sustained plasma levels over a period of 24 to 72 hours in mice (6). In contrast, the plasma half-lives in mice of topotecan and irinotecan are approximately 20 and 60 minutes, respectively (6, 18), see Supplementary Table S1 for a summary of relevant physicochemical properties. These agents are designed to produce longer systemic exposures but it is unclear how effectively they will access the tumor compartment and it has been theorized that they might benefit from an enhanced permeability and retention (EPR) effect, theoretically exploiting the greater permeability of tumor vessels for selective delivery of large agents. An additional unknown is that all camptothecin analogues studied undergo reversible, pH-dependent conversion from a closed lactone-ring conformation to an open carboxylate form (19, 20). The closed-ring form is known to be more potent but it is otherwise unclear what impact the two conformations, and their inherent effect on drug partitioning, may have on overall drug access and distribution within solid tumors (21, 22).
In this study, camptothecin autofluorescence was used to visualize drug localization within tumor cryosections that were subsequently immunostained using a multiplexed imaging procedure to identify tumor vasculature and perfusion status as well as tumor cell response. Relating drug exposure to activity on a microregional level over time made it possible to gain insight into the tumor microenvironmental-derived mechanisms involved in limiting drug activity. Two tumor cell lines, HCT116 (p53 wt) and HT29 (p53 mut), with differing sensitivity to camptothecins and p53 status, a key modulator of camptothecin response, were used (23–27). HT29 tumor xenografts have been characterized as having a more mature vascular phenotype as compared with the HCT116 model in terms of collagen and smooth muscle actin association with blood vessels (28). The two tumor types have been previously used to model drug penetration and were selected as they exhibited clearly demarcated regions under the influence of individual vessels and hence allowed for clearer interpretation of the interplay between microregional drug delivery and activity (29–32). In addition to the tumor-based studies, an in vitro 3D tissue disk assay was used. The tissue disks consist of tumor cells grown on a porous support membrane, which when grown to thicknesses more than approximately 150 μm, can effectively model the tumor extravascular compartment (12, 33–35). This assay permitted manipulation of pharmacokinetic properties via control of drug concentrations and duration, enabling a direct comparison between drug analogues with differing pharmacokinetic properties.
Materials and Methods
Monolayer and 3D tissue disk culture
HCT116 and HT29 cells were obtained from the ATCC and passaged for a maximum of 4 months. Mycoplasma status was tested on a monthly basis using Hoechst 33342 labeling of DNA. Cells were cultured in MEM (HyClone) with 10% FBS (Gibco/BRL) under 5% O2 and 5% CO2. Three-dimensional tissue disks were grown by seeding 106 cells into collagen-coated tissue culture inserts (CM 12 mm, pore size 0.4 μm; Millipore) and incubated for 16 hours to allow cells to attach to the porous membrane before being submerged in media and transferred to stirred growth vessels (32).
Mice and tumors
Female NOD.CB17-Prkdcscid/J mice were bred and maintained in our institutional animal facility in accordance with the Canadian Council on Animal Care guidelines. All experiments were approved by the Animal Care Committee of the University of British Columbia (Vancouver, British Columbia, Canada). Mice were allowed free access to standard laboratory rodent food and water and were used between 8 weeks and 14 weeks of age, typically weighing between 20 and 30 g. HCT116 cells (8 × 106 cells in 50 μL) or HT29 cells (5 × 106 cells in 50 μL) were implanted subcutaneously (s.c.) into the sacral region and tumors grown to 150 mm3 as assessed from caliper measurement of three orthogonal diameters (a,b,c) using the formula volume = π/6(abc). Mice were administered topotecan hydrochloride hydrate (Sigma) intraperitoneally at 10 mg/kg, irinotecan hydrochloride trihydrate (Hospira Canada) at 25 or 50 mg/kg i.v., or irinophore C (gift from Marcel Bally, Department of Experimental Therapeutics, BC Cancer Agency) also dosed at 50 mg/kg irinotecan i.v. (6). To label S-phase cells and tissue hypoxia, mice were administered 1,000 mg/kg bromodeoxyuridine (BrdUrd; Sigma-Aldrich) and 60 mg/kg pimonidazole by intraperitoneal injection 2 hours before tumor excision. To demarcate actively perfused blood vessels, mice were administered 35 μL of 0.6 mg/mL DiOC7(3) (Sigma-Aldrich) in 25% DMSO 5 minutes before tumor excision following which tumors were weighed and immediately frozen.
Camptothecin media stability and cellular accumulation
Cell culture media (MEM with 10% FBS) with or without HCT116 cells at 1% by volume were maintained under stirred conditions at 37 C, 5% CO2. Topotecan, irinotecan, or SN38 was added at 10 μmol/L and samples were taken over then next 4 hours. Samples from cell-free experiments were processed by diluting 1:10 in MeOH and then centrifuging at 16,000 × g for 15 minutes. For drug cellular accumulation experiments, two samples were taken at each time point, the first was processed as above, extracting drug from both media and cells, and the second was preprocessed to remove cells by centrifugation before processing the media alone (5 minutes at 3,000 g). Sample quantification was carried out on an Agilent 1260 Infinity HPLC using a mobile phase of 20% acetonitrile and 80% 50 mmol/L ammonium dihydrogen orthophosphate, pH 6.4, at 2 mL/minute, 40°C, and a Halo fused-core C18 column (50 × 4.6 mm, 2.7 μm, Advanced Materials Technology). Of note, 10 μL injections were made and detection of open- and closed-ring conformations of the camptothecins was done using fluorescence excitation 375 nm, emission 425 nm (irinotecan, SN38), and 525 nm (topotecan) modified from published methods (6, 36). Samples were processes and injected as they were being produced during the experiment. Closed-ring forms of irinotecan, SN38, and topotecan eluted at 4.5, 3.1, and 0.60 minutes, respectively, whereas open ring forms eluted at 0.78, 0.56, and 0.32 minutes and had relative fluorescence response of 81%, 117%, and 151% as compared with the closed-ring forms.
IHC A: Camptothecin, DiOC7(3), CD31, TUNEL, pimonidazole, and BrdUrd
Before immunostaining, slides were imaged for DiOC7(3) (ex: 488 nm, emission: 515–565 nm) to visualize vessel perfusion status and camptothecin (topotecan ex: 365 nm em: 515–565 nm; irinotecan and SN38 ex: 365 nm, em: 435–495 nm) to visualize drug location. Cryosections were then fixed in formalin for 15 minutes at room temperature. Vasculature was stained using a 1:500 dilution of hamster-anti-CD31 (clone 2H8) and 1:500 fluorescent Alexa 647 anti-hamster secondary (Invitrogen). Slides were imaged for CD31-positive vasculature and then processed to enable BrdUrd detection (citrate buffer, pH 6, at 120°C for 10 minutes at 20 psi; ref. 37). BrdUrd adducts were detected using a 1:500 dilution of monoclonal rat anti-BrdUrd (clone BU1/75) followed by 1:500 dilution of anti-rat Alexa 488 and hypoxia was detected via bound pimonidazole adducts using a 1:500 polyclonal rabbit–anti-pimonidazole (Hypoxyprobe) and a 1:500 Alexa 750 anti-rabbit secondary. Apoptosis was detected using a TUNEL kit (Roche Diagnostics) with a TMR red tagged dUTP. Cellular DNA was counter stained with Hoechst 33342.
IHC B: DiOC7(3), CD31, cleaved caspase-3, and pimonidazole
Before immunostaining, slides were imaged for DiOC7(3) and camptothecin tissue fluorescence as above and fixed in formalin. Vasculature was stained using a 1:500 dilution of rat-anti-CD31 (clone MEC 13.3) and 1:500 fluorescent Alexa 546 anti-rat secondary (Invitrogen). Apoptosis was detected using a 1:500 rabbit–anti-active caspase-3 antibody (BD 559565) and a 1:500 anti-rabbit Alexa 647 secondary. Hypoxia was then detected via bound pimonidazole adducts using a 1:500 mouse-anti-pimonidazole (Hypoxyprobe) and a 1:500 Alexa 488 anti-rabbit secondary. Cellular DNA was counter stained with Hoechst 33342.
The imaging system consists of a robotic fluorescence microscope (Zeiss Axioimager Z1), a cooled, monochrome CCD camera (Retiga 4000R; QImaging), a motorized slide loader and x–y stage (Ludl Electronic Products), and customized ImageJ software (public domain program developed at the U.S. National Institutes of Health, available at http://rsb.info.nih.gov/ij/) running on a Macintosh computer (Apple). The system allowed tiling of adjacent microscope fields of view. Using this system, images of entire tumor cryosections 1 to 3 cm2 were captured at a resolution of 1.5 μm/pixel.
Image analysis of tumor xenografts
Using ImageJ and user supplied algorithms, images of DiOC7(3), camptothecin-derived fluorescence, CD31, pimonidazole, and either cleaved caspase-3 or BrdUrd and TUNEL staining from each tumor section were overlaid and areas of necrosis and staining artifacts manually removed. CD31-positive regions were identified by selecting all pixels 10 SDs above the tissue background levels. CD31-positive regions that were less than 10 μm2 in size were considered artifacts and automatically removed from the analysis. CD31-positive objects that had at least 25% overlap by area with DiOC7(3) were counted as actively perfused. BrdUrd, TUNEL, and cleaved caspase-3–positive staining was identified by selecting pixels that were 5 SDs above tissue background levels. Measuring the distance from each point in the tissue to the nearest CD31-positive object and noting if BrdUrd, TUNEL, or cleaved caspase-3–positive was used to determine the relation between drug activity and distance to the nearest detected blood vessel. The data were tabulated so as to determine the fraction of positive pixels of the total number pixels found at each distance to a blood vessel. Camptothecin fluorescence was assessed via similar methods, using average signal intensity rather than the fraction of pixels above threshold.
One-way ANOVA tests were performed using Prism software (GraphPad). Significance of differences between multiple groups was compared using a Bonferroni post-test analysis.
Large disparity observed in microregional distribution of camptothecins in solid tumors
Figure 1 shows the typical pattern of drug fluorescence observed in HCT116 xenografts at 1 hour following treatment with topotecan (10 mg/kg i.p.), irinotecan (50 mg/kg i.v.), and irinophore C (50 mg/kg i.v.), reflecting typical doses used for the agents (6, 18). Images show drug-derived fluorescence for each compound along with blood vessel location and blood perfusion status at time of tumor excision. The images reveal that topotecan distributed within the tissue most efficiently, followed by irinotecan and finally irinophore C. Although topotecan exhibited build-up throughout the tissue, irinotecan seemed to only extravasate from some vessels, with many of the actively perfused, DiOC7(3) positive, vessels showing no drug in the surrounding tissue. Irinophore C, the liposomal formulation of irinotecan, showed the lowest level of extravasation and tissue distribution. In the HCT116 xenografts, irinotecan and irinophore C permeable vessels seemed to be randomly distributed throughout the tumor, whereas in the HT29 xenografts, the drug-permeable vessels were more commonly observed nearer the tumor margins, this despite an even distribution of perfused vessels throughout the tumors, as demarcated by the fluorescent dye DiOC7(3). Supplementary Figs. S1 and S2 show a comparison of the tissue distribution of irinotecan and irinophore C in whole HCT116 and HT29 tumor cryosections and Supplementary Fig. S3 compares irinotecan localization with that of vascular function imaged using DCE-MRI.
The time course of drug distribution in relation to tumor vasculature was followed to 72 hours post administration. Figure 2A shows a summary of drug deposition in HCT116 xenografts in relation to tumor vasculature and vessel patency over time. Topotecan can be seen to exhibit a relatively uniform distribution at 1 hour, followed by rapid wash-out over the next few hours and mostly gone by 8 hours, with fluorescence returned to control levels. These results were consistent with the relatively short plasma half-life of 0.3 hours in mice (18) that would result in a short period for drug accumulation. Irinotecan exhibited longer tissue retention with significant fluorescence still detectable at 4 hours and residual levels detectable out to 72 hours. The slower rate of wash-out of irinotecan, seen from the analysis, Fig. 2B, could potentially be due to greater retention of the drug compared with topotecan, suggested by its slower build-up in tissue, and also its longer plasma half-life of 0.8 to 1 hours. The nanoparticulate liposomal formulation of irinotecan and irinophore C exhibited the sharpest tissue gradient in relation to vasculature. Despite irinophore C's extended exposure, plasma half-life of 6 to 9 hours, its tissue distribution changed only slightly over 24 hours, indicating very limited tissue accumulation. Evidence for wash-out of irinophore C, a drop in drug-derived fluorescence, did not appear until 72 hours following administration. Figure 2B shows results from analysis of each drug in relation to tumor vasculature as averaged over entire tissue sections, up to 72 hours following their administration. At a depth of 100 μm away from vessels 1 hour drug levels reached 94% ± 5% (topotecan), 41% ± 10% (irinotecan), and 4.8% ± 2% (irinophore C) of the maximal levels seen near vessels.
Topotecan, irinotecan, and irinophore C, all uniformly inhibit proliferation at early time points but only irinophore C is able to maintain a sustained decrease in proliferation.
The impact of the camptothecins within solid tumor xenografts was assessed on a microregional scale in tissue cryosections using a multiplexed staining method, mapping tumor cell status in relation to distance from blood vessels. Proliferation was determined using the S-phase marker BrdUrd and apoptosis using both TUNEL and cleaved caspase-3 markers. Figure 3 summarizes findings in HCT116 xenografts following topotecan, irinotecan, and irinophore C treatment. Results show that despite exhibiting different tissue distributions, all three agents were able to completely inhibit proliferation throughout the tumors within 4 hours following drug administration, with no significant impact on tumor blood flow (see images Fig. 3A and proliferation analysis Fig. 3B). However, 24 hours following topotecan and irinotecan treatment, HCT116 tumors exhibited a rapid and full recovery to control proliferation levels; in contrast, irinophore C was able to produce a sustained inhibition for more than 1 week. No significant changes in microvessel density, vessel perfusion status, or tissue hypoxia were observed following treatment (see Supplementary Data and Supplementary Fig. S4 and S5).
The overall effects on proliferation and apoptosis in HCT116 and HT29 xenografts are summarized in Fig. 4. Results show that all three agents were able to completely inhibit proliferation at 4 to 8 hours following treatment in the HCT116 (p53 WT). In the HT29 (p53 mut) tumors, irinotecan produced marginal inhibition, whereas irinophore C was able to achieve a 50% to 75% reduction in proliferation, consistent with a reduced ability of the HT29 cells to enter p53-mediated cell-cycle block following drug-induced DNA damage. Proliferation was seen to approach control levels at 24 hours, following this a second dip in proliferation was consistently observed at 3 days following treatment. This second dip was attributed to a gap in cells ready to return to cycle due to the initial drug activity on the first day of treatment. Steady recovery to control levels was then observed between 4 days and 10 days following topotecan and irinotecan in the HCT116 xenografts. In contrast, irinophore C was able to maintain a sustained inhibition of proliferation out past 1 week, with only a partial recovery to 30% of control levels by 14 days. Apoptosis measured either with TUNEL or activated caspase-3 showed an initial surge 1 to 8 hours following topotecan and irinotecan treatment, whereas irinophore C displayed a delayed effect that increased out toward 7 to 14 days following treatment (Fig. 4, bottom).
SN38, active metabolite of irinotecan, exhibits fast tissue build-up compared with parent drug
Despite the significant differences in extravasation and distribution within the tumor compartment, the agents all exhibited a similar ability to inhibit proliferation throughout the tissue at early time points following treatment. One explanation for this would be if SN38, the active metabolite of irinotecan, was to distribute more efficiently than the parent drug. Tumor-based fluorescence experiments involving SN38 were not feasible due to the much lower drug levels of the metabolite relative to the parent drug so in vitro experiments were carried out to directly compare SN38 tissue penetration with that of irinotecan and topotecan using 3D tissue disks. In these experiments, HCT116 3D tissue disks grown to 300 μm in thickness were immersed in drug containing medium for 0.5, 1, 2, and 4 hours. Figure 5 summarizes the distribution of topotecan, irinotecan, irinophore C, and SN38-derived fluorescence in the 3D tissue disks. It was found that even at the earliest time point, 0.5 hour, SN38 exhibited a uniform distribution within the disks, matching topotecan, whereas irinotecan and irinophore C showed only limited tissue penetration at this point. Surprisingly, continuous drug exposure for 4 hours resulted in a counter-intuitive reduction in the fluorescence levels for all three drugs (see work comparing accumulation of open- vs. closed-ring conformations in the following section for an explanation). Camptothecin redistribution during washout was assayed by transferring tissue disks to drug-free medium following 1-hour exposure at 10 μmol/L drug. Results indicated that topotecan and SN38 experienced a rapid wash-out from tissue, whereas irinotecan showed a slower rate of removal. After 2 hours of wash-out, topotecan and SN38 were reduced to approximately 10% to 15% of initial levels, whereas irinotecan was still at approximately 50% of initial levels.
Irinotecan in the closed lactone-ring conformation accumulates significantly in cells
To investigate the decrease in tissue accumulation following the sustained 4-hour exposures in 3D disks, the stability and cellular accumulation of the agents were examined in stirred cell suspension. All three compounds topotecan, irinotecan, and SN-38 can exist in open or closed-ring forms and each maintains a reversible, pH-dependent, equilibration between the two forms. It was found that irinotecan exhibited a high degree of cellular accumulation in the closed-ring conformation but not the open-ring form and that this also occurred but to a lesser extent for topotecan and SN38. Figure 5C (left) shows the cellular accumulation of the drugs over time. Irinotecan initially reaches 20-fold higher concentrations in cells over the media levels but falls over time. Topotecan followed by SN38 are seen to exhibit a similar accumulation pattern and reversal but to lesser degrees. Figure 5C (right) shows the time course of conversion of closed-ring to open-ring conformation for each drug in cell-free media. Equilibration occurred with a half-life of approximately 45 minutes. These findings indicate that the limited distribution of irinotecan may be in part due to the high but transient accumulation in cells. The gradual improvement in irinotecan distribution over time can be explained by the conversion from the initially predominant closed-ring conformation to the nonaccumulating open-ring form.
Drug exposure and impact on tumor growth delay
The temporal patterns of proliferation in the tumors at early time points following treatment with irinotecan and topotecan suggest that repeat or split dosing of the drugs could result in more effective targeting of all cycling cells. To test this hypothesis, we conducted a series of tumor growth control experiments in which irinotecan or topotecan was given as single doses versus two split doses. Results, shown in Fig. 6A, indicate that splitting 50 mg/kg irinotecan into two 25 mg/kg doses separated by 24 hours yielded a modest improvement in growth control when given 2 weeks in a row. However, in comparison, a single dose of irinophore C yielded a much more substantial growth delay. Subsequent studies (Fig. 6B and D) further explored the efficacy of targeting out of cycle cells using dose-splitting strategies. In each case, treatment was given once per week for 3 weeks either as a single full dose or a split dose administered with a 2-hour gap or a 1-day gap. Results show that the split dose with a 1-day gap was most effective at controlling tumor growth. Irinotecan and topotecan both benefited from this strategy in HCT116 xenografts (Fig. 6B and C) but only marginal enhancement was seen in the HT29 xenografts (Fig. 6D). Overall these studies suggest that although targeting cells on two sequential days with irinotecan and topotecan improved growth control, these strategies were still outperformed by the extended plasma exposure provided by irinophore C.
The goal of this study was to directly visualize microregional localization and activity of camptothecins in solid tumors, and to identify the mechanisms determining tumor resistance and drug efficacy. The imaging of drug-derived fluorescence in relation to tumor vasculature identified the poor tissue distribution of irinotecan and irinophore C as potentially limiting their effectiveness. However, follow-up studies in 3D tissue disks showed that SN38, the active metabolite of irinotecan, exhibited much more efficient tissue penetration. Hence, the poor extravasation and tissue penetration observed for irinotecan and irinophore C were not likely to limit their efficacy, as SN38 is produced systemically following both these treatments. Consistent with this, the microregional activity of the agents showed that all three drugs, irinotecan, topotecan, and irinophore C were able to produce tumor wide inhibition of cell proliferation within 4 hours of treatment. The agents did differ significantly in the pattern of recovery that followed. Topotecan- and irinotecan-treated tumors recovered rapidly from the initial suppression of proliferation and by 24 hours, proliferation rates had returned to pretreatment levels, consistent with the rapid wash-out observed for both topotecan and SN38. Hence, lack of tissue retention seemed to be the most important factor limiting activity. In contrast, irinophore C induced a much more sustained inhibition of proliferation lasting 1 to 2 weeks and this improved performance was attributed to its longer systemic exposure (6). It was unclear what was the source of the delayed onset of apoptosis following irinophore C. Potentially its long exposure, which resulted in a sustained inhibition of proliferation and a delayed attempt at cell-cycle reentry might be at the root of this. Irinophore C and irinotecan have been reported to produce peak SN38 plasma levels of 5 μmol/L immediately following 50 mg/kg i.v. injection, with SN38 decreasing to 0.5 μmol/L by 4 hours in irinotecan-treated mice, whereas irinophore C-treated mice exhibited sustained SN38 levels of 2 μmol/L out to 24 hours (6).
On the basis of the tissue mapping studies, we hypothesized that the performance of irinotecan and topotecan could be improved by dose-splitting strategies in which drug would be administered 2 days in a row to target a larger number of cycling cells. This was along the lines of what had been found empirically for irinotecan and topotecan previously using a 5 days a week treatment schedule repeated for 2 weeks (15–17). In this study, tumor growth delay experiments showed that dose splitting between just 2 days in a week was an effective approach; however, the longer, sustained exposure produced by irinophore C was found to be better yet at controlling tumor growth.
Although the pattern of drug extravasation differed between the sensitive HCT116 and resistant HT29 tumor xenografts, it was concluded that this was not likely driving the differences in tumor response for these two cell types. The HCT116 tumors exhibited a random subset of actively perfused vessels that exhibited irinotecan/irinophore C extravasation. In contrast, the HT29 tumors displayed a predominance of drug-positive vessels around the periphery of the tumor, whereas perfused vessels in the core region showed less drug. In the case of the long lived and high-molecular weight irinophore C formulation, it is likely that differences in vessel leakiness were determining the observed patterns. Other possible factors such as variable blood flow and longitudinal vessel gradients would most likely be resolved by the long half-life of this drug. Because of the systemic release of the more uniformly distributing SN38, it was likely that these intertumor differences in drug distribution were not responsible for the observed differences in sensitivity to camptothecins and it was more likely that other known factors such as p53 status play the dominant role in determining tumor response of HCT116 versus HT29 cell types (23–27). Irinophore C did not seem to directly benefit from an EPR effect, and in general any increased tumor vessel permeability to the nanoparticles was confined to a subset all vessels and did not present a reliable way to access the tumor compartment as a whole.
A surprising finding of the study was that continuous exposure of tissue to camptothecins did not yield a steady build-up in drug levels. This seemed to be due to re-equilibration between closed and open-ring conformations of the drugs. Notably, irinotecan in the closed-ring form accumulated to a high degree in cells and did not distribute as uniformly within tissue compared with topotecan and SN38. Irinotecan in its closed-ring conformation accumulated 20-fold in cells and this was reduced to an approximately 4-fold accumulation after 4 hours over which time it had equilibrated to a predominantly open-ring conformation. Topotecan and SN38 followed a similar trend but to a much lesser degree in terms of cellular accumulation. The initial high level of tissue binding of irinotecan would impede its distribution but also help retain it during wash-out.
The poor tissue retention of the small molecule camptothecins, with a consequently rapid recovery of proliferation 24 hours following treatment, was identified as the key-limiting factor of irinotecan and topotecan efficiency in this study. This finding was supported by the greater efficacy observed for the high-molecular weight but longer acting irinophore C, and by the greater efficacy of dose-splitting strategies in the growth delay studies. These findings are consistent with those made during the preclinical development of topotecan and irinotecan, where it was found that treatment 5 days a week for 2 weeks was more effective than single dose strategies (15–17). In future work, it would be of interest to compare the effectiveness of dosing 2 or 3 days versus 5 days in a row each week. On the basis of the tumor mapping studies done here, the fraction of proliferating cells for the final days of a 2 week 5 day dosing schedule should be low, theoretically making the additional dosing less tumor specific and increasing systemic toxicity.
Evaluation of nanoparticle distribution over time suggested that in this case the EPR effect was not sufficient to aid in the activity of irinophore C and that it was the systemic release of irinotecan, and its conversion to SN38 over time was the key factor allowing irinophore C to exert activity throughout the tumors. Most nanodrugs and formulations will not benefit from this two-staged mechanism; hence, this study illustrates the need for careful characterization of the pattern of microregional activity of nanobased compounds. In regard to the camptothecins, findings of this study support the hypothesis that it is limited tissue retention that thwarts their effectiveness, and that designing analogues and formulations with longer plasma exposures that can access the tumor compartment will lead to more effective agents.
Disclosure of Potential Conflicts of Interest
No potential conflicts of interest were disclosed.
Conception and design: A.H. Kyle, J.H.E. Baker, M.J. Gandolfo, A.I. Minchinton
Development of methodology: A.H. Kyle, A.I. Minchinton
Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): A.H. Kyle, J.H.E. Baker, M.J. Gandolfo, A.I. Minchinton
Analysis and interpretation of data (e.g., statistical analysis, biostatistics, computational analysis): A.H. Kyle, J.H.E. Baker, M.J. Gandolfo, A.I. Minchinton
Writing, review, and/or revision of the manuscript: A.H. Kyle, J.H.E. Baker, S.A. Reinsberg, A.I. Minchinton
Study supervision: A.I. Minchinton.
This work was supported by the Canadian Institutes of Health Research operating grant MOP-86726 (to A.I. Minchinton), National Science and Engineering Research Council of Canada grant 327579-09 (to S.A. Reinsberg), Canadian Cancer Society Research Institute grant 20251 (to S.A. Reinsberg and J.H.E. Baker), and Canadian Breast Cancer Foundation Postdoctoral Fellowship, BC/Yukon Region (to J.H.E. Baker).
The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.