Zheng JH, Chen CT, Au JL and Wientjes MG Time- and Concentration-Dependent Penetration of Doxorubicin in Prostate Tumors AAPS PharmSci 2001; 3 (2) article 15 (https://www.pharmsci.org/scientificjournals/pharmsci/journal/01_15.html).

Time- and Concentration-Dependent Penetration of Doxorubicin in Prostate Tumors

Submitted: September 25, 2000; Accepted: May 7, 2001; Published: May 15, 2001

Jenny H. Zheng^1,2, Chiung-Tong Chen^1,3, Jessie L.-S. Au¹ and M. Guill Wientjes¹

¹College of Pharmacy and James Cancer Hospital and Solove Research Institute, The Ohio State University, Columbus, OH

²Food and Drug Administration, 9201 Corporate Blvd, HFD-880, Rockville, MD 20850

³Division of Biotechnology and Pharmaceutical Research, National Health Research Institutes, 1F, Building C, 103, Lane 169, Kang-Ning Street, Hsi-Chih, Taipei 221, Taiwan

Abstract

The penetration of paclitaxel into multilayered solid tumors is time- and concentration-dependent, a result of the drug-induced apoptosis and changes in tissue composition. This study evaluates whether this tissue penetration property applies to other highly protein-bound drugs capable of inducing apoptosis. The penetration of doxorubicin was studied in histocultures of prostate xenograft tumors and tumor specimens obtained from patients who underwent radical prostatectomy. The kinetics of drug uptake and efflux in whole tumor histocultures were studied by analyzing the average tumor drug concentration using high-pressure liquid chromatography. Spatial drug distribution in tumors and the drug concentration gradient across the tumors were studied using fluorescence microscopy. The results indicate that drug penetration was limited to the periphery for 12 hours in patient tumors and to 24 hours in the more densely packed xenograft tumors. Subsequently, the rate of drug penetration to the deeper tumor tissue increased abruptly in tumors treated with higher drug concentrations capable of inducing apoptosis (i.e., = 5 µm), but not in tumors treated with lower concentrations. These findings indicate a time- and concentration-dependent penetration of doxorubicin in solid tumors, similar to that of paclitaxel. We conclude that doxorubicin penetration in solid tumors is time- and concentration-dependent and is enhanced by drug-induced cell death.

Introduction

Drug delivery to a solid tumor is governed by several factors that differ for systemic and regional treatments. Following a systemic intravenous injection, drug delivery to the tumor core involves 3 processes (ie, distribution through vascular space, transport across microvessel walls, and diffusion through interstitial space in tumor tissue). ¹ When the drug is directly injected into a tumor, such as by intratumoral injection or by direct instillation into peritumoral space as in intravesical therapy of superficial bladder cancer and in intraperitoneal dialysis of ovarian cancer, drug delivery to tumor cells is primarily by diffusion through interstitial space.^2-6

The inability of a drug to penetrate a solid tumor is considered resistance of solid tumors to anticancer drugs.^7-11 For example, penetration of doxorubicin in 3-dimensional tumor cell spheroids after 1 to 2 hours is limited to the periphery.^{7-9 12} Similarly, a steep concentration gradient in breast tumors has been observed in patients.¹³ Hence, a better understanding of the determinants of drug penetration into solid tumors is needed.

We recently studied the penetration of paclitaxel, a highly protein-bound drug, in solid tumors. The study was performed under in vitro conditions where paclitaxel was placed in the culture medium surrounding histocultures of tumor fragments (~1 mm³ ). The results show that a high tumor cell density is a barrier to paclitaxel penetration in tumor tissue; paclitaxel penetration is restricted to the tumor periphery until the cell density is reduced as a result of paclitaxel-induced apoptosis, at which time paclitaxel distributes evenly throughout the tumor.¹⁴ This study examined whether the time-dependent and apoptosis-enhanced drug delivery applies to other drugs. We examined several aspects of doxorubicin penetration into solid tumors (ie, kinetics of drug penetration and effects of tumor cell density and tissue composition on drug penetration). Doxorubicin, similar to paclitaxel, is a highly protein-bound drug. The study was performed using histocultures of prostate tumors obtained from patients and human xenograft tumors maintained in immunodeficient mice. Histocultures are fragments of tumors obtained from a human or animal host, cut to approximately 1 mm³ and cultured on a collagen matrix. Histocultures maintain a 3-dimensional structure and, therefore, cell-cell interaction and clonal heterogeneity. This is similar to spheroids, which are aggregates of cultured tumor cells, sometimes cocultured with fibroblasts. ^{15 16} We preferred using histocultures for the current study because this allows us to study drug penetration in patient tumor material and because the results obtained using patient tumors are more likely to be clinically relevant. The presence of stromal cells and matrix material is considered important for prostate tumor growth¹⁷ and, as shown in this study, plays a role in the penetration and accumulation of doxorubicin in prostate tumor. In addition, the clinical relevance of the histoculture system has been demonstrated in retrospective and semiprospective preclinical and clinical studies; drug response in human tumor histocultures correlates with chemosensitivity and survival of cancer patients to several chemotherapeutic drugs.^18-20

Materials and Methods

Chemicals and Supplies

Doxorubicin and epirubicin were gifts from Pharmacia & Upjohn (Milan, Italy; Albuquerque, NM) or purchased from Sigma Co (St Louis, MO). Male athymic BALB/C Nu/Nu mice were purchased from the National Cancer Institute (Frederick, MD); cefotaxime sodium from Hoechst-Roussel Inc (Somerville, NJ); gentamicin from Solo Pak Laboratories (Franklin Park, IL); fetal bovine serum (FBS), nonessential amino acids, L-glutamine, minimum essential medium (MEM), Dulbecco's Modified Eagle Medium (DMEM), and RPMI 1640 medium from GIBCO Laboratories (Grand Island, NY); sterile pigskin collagen gel (Spongostan standard) from Health Designs Industries (Rochester, NY); cryotome imbedding polymer from Miles Inc (Ellchart, IN); solid phase extraction tubes (Supelclean LC-18) from Supelco (Bellefonte, PA); a rotor-stator type of tissue homogenizer (Tissumizer) from Tekmar (Cincinnati, OH); and Pecosphere reversed-phase C₁₈ columns (3 µm particle size, 83 mm x 4.6 mm) from Perkin-Elmer (Norwalk, CT).

Tumor procurement

Surgical specimens of human prostate tumors were obtained through the Tumor Procurement Service at The Ohio State University Comprehensive Cancer Center from patients who underwent radical prostatectomy. Tumor specimens were placed in MEM within 10 to 30 minutes after surgical excision, stored on ice, and prepared for culturing within 1 hour after excision.

Human prostate tumor xenografts maintained in nude mice

The two human prostate xenograft tumors (ie, the androgen-dependent CWR22 and the androgen-independent PC3 tumors) were established and maintained as described previously.^21-23 Briefly, minced tumor tissue was mixed with an equal volume of Matrigel, and 0.3 mL of the mixture was implanted into both flanks of a mouse. Tumors were harvested when they reached a size of 1 g at about 7 weeks for CWR22; 4 weeks for PC3. For the CWR22 tumor, animals were implanted subcutaneously with a testosterone pellet (12.5 mg/tablet, Innovative Research of America, Toledo, OH) 3 days before tumor implantation.

Histocultures

Patient prostate or tumor xenograft specimens were processed as previously described. ¹⁴ Briefly, specimens were washed 3 times and dissected into fragments measuring about 1 mm³ . The culture medium consisted of MEM/DMEM (1:1) for patient tumors, or RPMI 1640 for PC3 xenograft tumor, supplemented with 9% heat-inactivated FBS, 2 mM l-glutamine, 0.1 mM nonessential amino acids (only for MEM/DMEM), 90 µg/ml gentamicin and 90 µg/mL cefotaxime sodium. For the CWR22 xenograft tumor, the culture medium consisted of a 1:1 mixture of MEM and DMEM, 10% fetal bovine serum, 0.1 mM nonessential amino acids, 2 mM l-glutamine, 1 mM sodium pyruvate, MEM vitamin solution, and 40 µg/mL gentamicin.

Drug uptake and efflux in histocultures

Histocultures were placed on a 1-cm² piece of presoaked collagen gel (5 histocultures per gel) and incubated with 4 mL culture medium in 6-well plates. The culture medium was refreshed every other day. After 3 to 4 days, the histocultures were treated with 0.02 to 20 µM doxorubicin for up to 96 hours. We have shown that these concentrations were sufficient to inhibit proliferation and induce cell death in patient prostate tumors. ²³ For the efflux study, tumor histocultures were incubated with doxorubicin for 96 hours. The drug-containing medium was then exchanged with drug-free medium, and histocultures and aliquots of medium were collected at predetermined times. The histocultures were blot-dried on filter paper and weighed.

For each tumor, 3 to 5 histocultures were used for each concentration and each time point. The study design of experiments using patient tumors was dictated by the size of the specimens. On some occasions, specimens from an individual patient were only sufficient to study drug uptake and efflux at 1 or more, but not all, drug concentrations. Ten patient tumors were used. For the xenograft tumors, specimens from individual animals were sufficiently large that each tumor was used for studying uptake and efflux at all drug concentrations.

HPLC analysis of doxorubicin concentration

The concentration of doxorubicin in culture medium was analyzed by high-pressure liquid chromatography (HPLC); the concentration of doxorubicin in tumors was analyzed by HPLC and quantitative fluorescence microscopy. Drug concentration in tissue was calculated as (drug amount) divided by (tissue weight) and was expressed in molar terms.

For HPLC analysis, we used previously published methods^{24 25} with minor modifications, as follows. Epirubicin was used as the internal standard and was added before sample extraction. For tumor histocultures, samples (average weight of ~5 mg) were homogenized for 1 minute with 2 mL acidified methanol (5% of 50 mM potassium phosphate buffer [pH 3.0] in methanol). Homogenates adhering to the homogenizer were recovered by rinsing with 3 mL of methanol. The methanolic extract was reduced to a volume of less than 2 mL by evaporation, followed by mixing with 3 mL of 10 mM potassium phosphate, pH 8.0, and 2 mL methanol. The final mixture was loaded on a C18 solid phase extraction column, which was preconditioned with 3 mL 100% methanol, followed by 3 mL of a 1:3 mixture of methanol:20 mM potassium phosphate, pH 8.0. After washes with 1 mL of water followed by 2 mL of 50% methanol in water, the analytes were eluted with 6 mL of a 95:5 mixture of methanol:50 mM potassium phosphate, pH 3.0. The extract was then evaporated to dryness. The residue was reconstituted with the mobile phase and analyzed by HPLC.

For the analysis of doxorubicin in culture medium, proteins in the culture medium were precipitated with 10% trichloroacetic acid, and the supernatant obtained after centrifugation at 7000 g for 10 minutes was analyzed by HPLC.

The reversed-phase isocratic HPLC analysis was performed using a Pecosphere C18 column and a mobile phase of 30% acetonitrile in 20 mM potassium phosphate (pH 3.0), at a flow rate of 0.8 mL/min. Doxorubicin and epirubicin were detected with the use of a scanning fluorescence detector. The excitation and emission wavelengths were 480 nm and 550 nm, respectively. The retention time was approximately 7 minutes for doxorubicin and approximately 9 minutes for epirubicin. Standard curves were linear within the range of 1 ng/mL to 100 ng/mL (ie, 0.002 µM to 0.17 µM). Samples of culture medium containing high doxorubicin concentrations were diluted as needed.

Microscopic evaluation of doxorubicin penetration and distribution in tumors

Spatial distribution of doxorubicin in tumor tissue was visualized using fluorescence microscopy. Histocultures were washed twice by dipping them in ice-cold drug-free medium, blot-drying, mounting on cryostat chucks with embedding matrix, placing them in a cryostat at -20°C, and cutting them into 10-µm sections. Sections were thaw-mounted on glass microscope slides and heat-fixed on a slide warmer for 15 minutes at 30°C. Slides were then covered with a coverslip, sealed with rubber cement, and evaluated using fluorescence microscopy with excitation and emission wavelengths of 546 nm and 565 nm, respectively. The captured fluorescence images were analyzed using Optimas image analysis software (Silver spring, MD). At least 3 readings were obtained for each data point.

To establish the standard curves (each curve contained 6 data points) for measuring doxorubicin concentration in tissues by fluorescence microscopy, a doxorubicin solution (2 µL) was applied to microscopic sections of blank dog prostate tissue (10 µm thick), to cover a surface area of approximately 1.5 cm² to 2 cm² . The average fluorescence intensity per area was measured and plotted against the applied doxorubicin concentrations to obtain the standard curves. When analyzing the doxorubicin in the actual samples, at least 3 sections were used per tumor and at least 3 tumors were used per time point.

Cell density

After tissue slides had been analyzed by fluorescence microscopy, the coverslips were removed and the slides stained with hematoxylin and eosin. The histologic images were captured using light microscopy and the cell densities were quantified using the Optimas software. Only cells with a diameter larger than 4 µm were counted.

Data analysis

Differences in mean values between groups were analyzed using the Student t test by SAS (Cary, NC).

Results

Accumulation and retention of doxorubicin in CWR22 tumor histocultures

The accumulation and retention of doxorubicin in tumors was measured using HPLC, which specifically detects unchanged doxorubicin. The doxorubicin concentrations in the CWR22 tumor increased with time and reached plateau levels between 48 and 96 hours (Figure 1 ,Table 1 ). The maximum drug concentration in tumors increased with the initial drug concentration in the culture medium; the ratio of the maximal tumor concentration to the final concentration in the medium was about 100. The fractions of drug concentration remaining in tumors after 24 and 48 hours were about 60% and 40%, respectively. The high drug accumulation in the tumors and the slow drug release from the tumors are likely the result of the drug binding to intracellular macromolecules. ^8-9

Similarities and differences in doxorubicin penetration and accumulation in patient and PC3 xenograft tumor

We compared the uptake of doxorubicin in patient and xenograft tumors. These tumors displayed different tissue composition and structure, which, as shown following, are important determinants of drug penetration. Because of the slow growth of the CWR22 tumor (1 g in ~7 weeks), the subsequent studies used the more rapidly growing PC3 tumor (1 g in ~4 weeks). Fluorescence microscopy was used to study the doxorubicin penetration and the intratumor concentration gradient as a function of the depth of drug penetration. Because the HPLC analysis of tumor homogenates did not detect the presence of doxorubicin metabolites (data not shown), the fluorescence intensity represented unchanged doxorubicin. The average width of the cross section of the histocultures was between 600µm and 800 µm, or about 60 to 100 cell-layers thick. We measured the drug penetration from the periphery (25 µm, referred to as periphery of tumor) to 325 µm, which represents the central region of the tumor (referred to as center of tumor).

Figure 2 shows the fluorescence micrographs. For both patient and xenograft tumors, drug accumulation in the periphery and the center of a tumor increased with time. At the lower concentration (ie, 1 µM), doxorubicin remained in the periphery of both patient and xenograft tumors at 72 hours. At higher concentrations (ie, 5 µM or 20 µM), doxorubicin was initially confined to the periphery (12 hours for patient tumors and 24 hours for xenograft tumors), followed by an abruptly enhanced drug penetration such that even distribution in histocultures was attained shortly after (24 hours for patients tumors and 36 hours for xenograft tumors).

Figure 3 shows the doxorubicin concentrations as a function of time and drug concentration. The data are presented as tumor-to-medium concentration ratios to standardize for the time-dependent changes in the extracellular drug concentration. Hence, a constant tumor-to-medium concentration ratio across the tumor indicates the attainment of equilibrium. Conversely, a declining concentration ratio from the periphery to the center indicates that the equilibrium was not achieved. For both patient and xenograft tumors, the concentration gradient from the periphery to the center of a tumor decreased with increasing treatment time and with increasing drug concentration. The periphery-to-center concentration gradient was the highest at the lowest initial extracellular concentration of 1 µM and decreased with increasing extracellular concentration, resulting in concentrations in the center being approximately equal to the concentrations in the periphery after treatment with 5 µM and 20 µM for at least 24 hours. Compared to the patient tumors, the xenograft tumor showed greater periphery-to-center concentration gradients-as indicated by the steeper concentration decline over tissue depth-for all 3 initial extracellular concentrations. The time required to reduce the periphery-to-center concentration gradient to 0, after treatment with 5 µM and 20 µM doxorubicin, was shorter in the patient tumors than in the xenograft tumor (ie, 24 hours versus 36 hours). The differences between patient and xenograft tumors, as shown following, result from the differences in tissue composition and structure.

Effect of tumor cell density on doxorubicin penetration and accumulation in tumor

The previous data indicate a delay in doxorubicin penetration to the center of the tumor; and a longer delay in xenograft tumors than in patient tumors. We have shown that, for paclitaxel, this delay is not the result of drug diffusion from culture medium to histocultures, but is the result of a high tumor-cell density¹⁴ ; a higher cell density corresponds with a smaller fraction of interstitial space. This results in increased tortuosity of the interstitial diffusion channels and slower drug diffusion.^{26 27} Similar findings were observed in the present study. Figure 4 shows that the xenograft tumor contained more tightly packed tumor cells, fewer stromal cells, and less interstitial space compared to patient tumors, which is consistent with the longer delay in doxorubicin penetration in xenograft tumors. The cell density in untreated histocultures of xenograft tumors was significantly higher than the density in patient tumors (2418 ± 66 versus 1864 ± 25 cells/mm² , P < .05).

In both xenograft and patient tumors, we observed a higher fluorescence intensity in cells compared with interstitial space. This observation, together with the observation of the extensive doxorubicin accumulation in tumor cells (ie, 100X the extracellular concentration; see Table 1 ), suggest the difference in cell density in patient and xenograft tumors as the cause of the difference in their drug accumulation. The magnitude of the difference in cell density between the xenograft and patient tumors (23%) is within the range of the difference of drug accumulation between these tumors (13% to 33%).

Effect of drug-induced cell death on doxorubicin penetration in tumor

The abrupt change in the rate of doxorubicin penetration to the center of a tumor was observed only at the higher concentrations of 5 µM and 20 µM and not at 1 µM (Figures 2 and 3 ). The 5-µM and 20-µM concentrations were near to or exceeded the drug concentration required to produce 50% cell death for a 96-hour treatment ²³ , whereas 1 µM was below this concentration. Figure 4 shows the reduction of tumor cell density over time after treatment with 20 µM doxorubicin, although no change was observed after treatment with 1 µM doxorubicin. This suggests that the abrupt change in drug penetration that occurred only after treatment with high drug concentrations is the result of drug-induced cell death and reduction of cell density. This is further supported by the inverse correlation between the average tumor concentration and the tumor cell density in the periphery of the xenograft tumor after treatment with 20 µM doxorubicin (Figure 5 ).

Discussion

This study's results indicate that the rate of doxorubicin penetration to the center of the tumor depends on initial extracellular drug concentration, treatment time, and tumor type (ie, tissue composition). Drug penetration was faster at higher concentrations, and the effect of concentration on the penetration rate was more significant in tumors with high tumor-cell density than in tumors with low density. A minimum treatment time of 24 to 36 hours was required for doxorubicin to penetrate a depth of 300 µm. Our results also indicate that the extent of maximal doxorubicin accumulation in tumor cells in a solid tumor depends on initial extracellular drug concentration, treatment time, tumor type, and tumor cell density.

Our finding that the penetration of doxorubicin in a solid tumor is confined to the periphery in the first 12 hours is consistent with the findings in tumor spheroids.^7-9 The slow penetration of doxorubicin is considered characteristic for high molecular weight molecules as a result of its extensive binding to proteins.^{1 9} The finding that a high tumor-cell density reduced doxorubicin penetration in a solid tumor is consistent with an earlier finding that coating of Teflon membranes with a multilayer of tumor cells (2-4 ± 106 cells, 200 µm thick) reduces the transmembrane transport of doxorubicin by more than 90%.¹¹

The two mechanisms often implicated in the slow penetration of drugs into solids tumors are the impeded influx resulting from high oncotic pressure ²⁸ or the drug efflux by the mdr1 p-glycoprotein. ²⁹ But neither of these mechanisms could be the cause of the slow doxorubicin penetration in the PC3 histocultures because PC3 cells do not express Pgp ^{30 31} and because histocultures lack the capillary blood flow needed to supply the oncotic pressure.

When compared with patient tumors, the higher density of epithelial cancer cells in xenograft tumors correlates with a slower drug penetration rate and a higher drug accumulation. These data suggest cellularity as a major determinant of the rate and extent of doxorubicin penetration and accumulation in solid tumors. Qualitatively, these findings are identical to our previous observations on the paclitaxel penetration and accumulation in solid tumors.¹⁴ As shown earlier for paclitaxel, apoptosis is required for enhanced drug penetration. Under conditions in which either insufficient drug concentration or insufficient time for apoptosis occurs, no enhancement in paclitaxel penetration was observed.³² Hence, enhanced drug penetration in solid tumors caused by drug-induced cell death appears a common phenomenon for at least 2 highly protein-bound drugs (ie, paclitaxel and doxorubicin). Our observations further demonstrate an interesting new concept in the relationship between drug delivery and drug effect. In addition to the general belief that drug delivery to tumor cells determines the antitumor activity, our finding indicates that the pharmacological effect of a drug can modify its delivery.

The finding that drug-induced apoptosis resulted in enhanced drug penetration in solid tumors may have clinical implications. We recently completed a study investigating the effect of dosing regimens on delivery of paclitaxel to solid tumors. The results showed that an initial apoptosis-inducing loading dose, followed by a second dose administered when apoptosis had occurred, resulted in a 50% higher drug concentration in tumors as compared with other treatment schedules where either the dose intensity was not sufficient to induce apoptosis or the dosing intervals were not sufficiently long for apoptosis to take place.³² It is noted that the doxorubicin concentrations used to induce apoptosis in the PC3 histocultures exceeded the clinically achievable concentrations (i.e., 5 µM versus 200 nM).^33-35 However, the 200 nM doxorubicin concentration is sufficient to induce cell death in histocultures of patient tumors.²³ Additional in vivo studies, such as those described for paclitaxel ³² , are needed to determine whether the tumor delivery of doxorubicin can be enhanced by manipulating the dosing schedule.

Conclusion

In summary, our results indicate drug-induced cell death as a key determinant of the rate and extent of doxorubicin penetration in solid tumors. The delivery of doxorubicin to cells in a solid tumor is a dynamic process determined by both the drug concentration and the treatment duration and the usual processes involved in drug transport (ie, distribution through vascular space, transport across microvessel walls, and diffusion through interstitial space in tumor tissue). That the pharmacological effects of doxorubicin affect its delivery will need to be taken into consideration when designing treatment schedules to maximize the drug delivery to the hard-to-reach tumor cells distant from the vasculature or from a regional delivery site. For example, a treatment schedule to include a pretreatment to induce cell death may enhance drug delivery to such sites.

Acknowledgements

This study was partly supported by research grant R01CA74179 from the National Cancer Institute, NIH, DHHS. The Tumor Procurement Service was partly supported by Cancer Center Support Grant P30CA16058 from the National Cancer Institute, NIH, DHHS. Dr. Zheng was partly supported by a Pharmacia-Upjohn Fellowship.

References

1. Jain RK. Barriers to drug delivery in solid tumors. Sci Am. 1994;271:58-65. [PUBMED]

2. Nativ O, Aronson M, Medalia O, et al. Anti-neoplastic activity of paclitaxel on experimental superficial bladder cancer: in vivo and in vitro studies. Int J Cancer. 1997;70:297-301. [PUBMED]

3. Song D, Wientjes M G, Au J L-S. Bladder tissue pharmacokinetics of intravesical taxol. Cancer Chemother Pharmacol. 1997;40:285-292. [PUBMED]

4. Song D, Wientjes MG, Gan Y, Au JL-S. Bladder tissue pharmacokinetics and antitumor effect of intravesical 5-fluorouridine. Clin Cancer Res. 1997;3:901-909. [PUBMED]

5. Markman M, Francis P, Rowinsky E, Hoskins W. Intraperitoneal paclitaxel: a possible role in the management of ovarian cancer? Semin Oncol. 1995;22:84-87. [PUBMED]

6. Markman M. Intraperitoneal therapy of ovarian cancer. Semin Oncol. 1998;25:356-360. [PUBMED]

7. Kerr DJ, Kaye SB. Aspects of cytotoxic drug penetration, with particular reference to anthracyclines. Cancer Chemother Pharmacol. 1987;19:1-5. [PUBMED]

8. Durand RE. Slow penetration of anthracyclines into spheroids and tumors: a therapeutic advantage? Cancer Chemother Pharmacol. 1990;26:198-204. [PUBMED]

9. Erlanson M, Daniel-Szolgay E, Carlsson J. Relations between the penetration, binding and average concentration of cytostatic drugs in human tumour spheroids. Cancer Chemother Pharmacol. 1992;29:343-353. [PUBMED]

10. Baguley BC, Finlay GJ. Pharmacokinetic/cytokinetic principles in the chemotherapy of solid tumours. Clin Exp Pharmacol Physiol. 1995;22:825-828. [PUBMED]

11. Tunggal JK, Cowan DS, Shaikh H, Tannock IF. Penetration of anticancer drugs through solid tissue: a factor that limits the effectiveness of chemotherapy for solid tumors. Clin Cancer Res. 1999;5:1583-1586. [PUBMED]

12. Durand RE. Distribution and activity of antineoplastic drugs in a tumor model. J Natl Cancer Inst. 1989;81:146-152. [PUBMED]

13. Lankelma J, Dekker H, Luque FR, et al. Doxorubicin gradients in human breast cancer. Clin Cancer Res. 1999;5:1703-1707. [PUBMED]

14. Kuh HJ, Jang SH, Wientjes MG, Weaver JR, Au JL-S. Determinants of paclitaxel penetration and accumulation in human solid tumor. J Pharmacol Exp Ther. 1999;290:871-880. [PUBMED]

15. Hamilton G. Multicellular spheroids as an in vitro tumor model. Cancer Lett. 1998;131:29-34. [PUBMED]

16. Kunz-Schughart LA. Multicellular tumor spheroids: intermediates between monolayer culture and in vivo tumor. Cell Biol Int. 1999;23:157-161. [PUBMED]

17. Kabalin JN, Peehl D M, Stamey TA. Clonal growth of human prostatic epithelial cells is stimulated by fibroblasts. Prostate. 1989;14:251-263. [PUBMED]

18. Robbins KT, Connors KM, Storniolo AM, Hanchett C, Hoffman RM. Sponge-gel-supported histoculture drug-response assay for head and neck cancer. Correlations with clinical response to cisplatin. Arch Otolaryngol Head Neck Surg. 1994;120:288-292. [PUBMED]

19. Furukawa T, Kubota T, Hoffman RM. Clinical applications of the histoculture drug response assay. Clin Cancer Res. 1995;1:305-311. [PUBMED]

20. Kubota T, Sasano N, Abe O, et al. Potential of the histoculture drug-response assay to contribute to cancer patient survival. Clin Cancer Res. 1995;1:1537-1543. [PUBMED]

21. Pretlow TG, Wolman SR, Micale MA, et al. Xenografts of primary human prostatic carcinoma. J Natl Cancer Inst. 1993;85:394-398. [PUBMED]

22. Nagabhushan M, Miller CM, Pretlow TP, et al. CWR22: the first human prostate cancer xenograft with strongly androgen-dependent and relapsed strains both in vivo and in soft agar. Cancer Res. 1996;56:3042-3046. [PUBMED]

23. Chen CT, Au JL-S, Wientjes MG. Pharmacodynamics of doxorubicin in human prostate tumors. Clin Cancer Res. 1998;4:277-282. [PUBMED]

24. Chai M, Wientjes MG, Badalament RA, Burgers JK, Au JL-S. Pharmacokinetics of intravesical doxorubicin in superficial bladder cancer patients. J Urol. 1994;152:374-378. [PUBMED]

25. Cox SK, Wilke AV, Frazier D. Determination of adriamycin in plasma and tissue biopsies. J Chromatogr. 1991;564:322-329. [PUBMED]

26. Schultz JS, Armstrong W. Permeability of interstitial space of muscle (rat diaphragm) to solutes of different molecular weights. J Pharm Sci. 1978;67:696-705. [PUBMED]

27. Fox JR, Wayland H. Interstitial diffusion of macromolecules in the rat mesentery. Microvasc Res. 1979;18:255-276. [PUBMED]

28. Stohrer M, Boucher Y, Stangassinger M, Jain RK. Oncotic pressure in solid tumor is elevated. Cancer Res. 2000; 60:4251-4255. [PUBMED]

29. Kaye SB. Multidrug resistance: clinical relevance in solid tumours and strategies for circumvention. Curr Opin Oncol. 1998; 1:S15-S19. [PUBMED]

30. Chen CT, Au JL-S, Wientjes MG. Androgen-dependent and -independent human prostate xenograft tumors as models for drug activity evaluation. Cancer Res. 1998;58:2777-2783. [PUBMED]

31. Van Brussel JP, van Steenbrugge GJ, Romijn JC, Schroder, FH, Mickisch GH. Chemosensitivity of prostate cancer cell lines and expression of multidrug resistance-related proteins. Eur J Cancer. 1999;35:664-671. [PUBMED]

32. Jang S-H, Wientjes MG, Au JL-S. Enhancement of paclitaxel delivery to solid tumors by apoptosis-inducing pretreatment: effect of treatment schedule. J Pharmacol Exper Therap. 2001; in press.

33. Legha SS, Benjamin RS, Mackay B, et al. Reduction of doxorubicin cardiotoxicity by prolonged continuous intravenous infusion. Ann Intern Med. 1982;96:133-139. [PUBMED]

34. Speth PAJ, Linssen PCM, Holdrinet RSG, Haanen C. Plasma and cellular adriamycin concentrations in patients with myeloma treated with ninety-six-hour continuous infusion. Clin Pharmacol Ther. 1987;41:661-665. [PUBMED]

35. Bugat R, Robert J, Herrera A, et al. Clinical and pharmacokinetic study of 96-h infusions of doxorubicin in advanced cancer patients. Eur J Cancer Clin Oncol. 1989;25:505-511. [PUBMED]