Doxorubicin Gradients in Human Breast Cancer

Jan Lankelma,1 Henk Dekker,Rafael Fernandez Luque, Sylvia Luykx,Klaas Hoekman, Paul van der Valk,Paul J. van Diest, and Herbert M. PinedoDepartments of Medical Oncology [J. L., H. D., R. F. L., S. L., K. H.,H. M. P.] and Pathology [P. v. d. V., P. J. v. D.], Free UniversityHospital, 1007MB Amsterdam, the Netherlands

ABSTRACTTen patients with locally advanced breast cancer were

given doxorubicin i.v., and an incision biopsy was subse-quently taken. Doxorubicin autofluorescence was examinedusing computerized laser scanning microscopy, and mi-crovessels were immunostained in the same sections. Over-lays of both pictures revealed doxorubicin gradients in tu-mor islets with high concentrations in the periphery and lowconcentrations in the center of the tumor islets. Gradientswere most pronounced shortly after the injection, but theycould still be detected 24 h later. No gradients were observedin connective tissue. This study demonstrates a serious riskof the drug not reaching all of the cancer cells in those casesin which the cancer cells are densely packed in islets. Theefficacy of drug treatment will thus depend on the histologyof the tumor tissue.

INTRODUCTIONEpithelial cancer is currently the most common cause of

cancer-related death. Although the normal architecture ofintact epithelium is lost, in epithelial cancer, the frequentfinding of desmosomes and occasionally tight junctions (1)indicate that the malignant epithelium may at least partiallyretain its function as a barrier,e.g., against drugs. A lowdensity of blood capillaries may limit tumor growth (2)because of reduced delivery of oxygen and nutrients. How-ever, during chemotherapy, reduced delivery of anticancerdrugs may mean a growth advantage for the most remotetumor cells. Drug transport out of the blood capillaries andinto the tissue can occur through the cells (transcellularpathway) and through the interstitium (paracellular pathway;Ref. 3). Doxorubicin is one of the most effective agents forthe treatment of epithelial cancer. However, the drug rarelyeliminates all cancer cells in patients with advanced disease,

including breast cancer. Previously,in vitro experimentsusing multicell spheroids have demonstrated doxorubicingradients (4, 5). This study was undertaken to investigatewhether such gradients can be foundin vivo as well.

At the cellular level, P-glycoprotein (6–8) and multidrugresistance protein (9, 10) can mediate drug resistance by de-creasing transport to the target. Drug pumping by these proteinsacross the plasma membrane resulted in lower intracellular drugconcentrations (11–14). At the tissue level, drug transport bar-riers have been recognized for large molecules (15, 16), which,because of their low diffusion rate, depend largely on convectiveflows for their transport. Furthermore, they may also be ham-pered by the high hydrostatic pressure outside the microvessels,conditions that often occur in solid tumors (15, 16). The nettransport of small molecules (such as most drugs with aMr ,1000) by diffusion may slow down as a result of binding torelatively large molecules that move at a much slower speed(17). Moreover, when the transport rate across the plasma mem-brane is low in combination with strong intracellular binding,drug transport through the cells can be slowed down even moredrastically. Then, just after i.v. injection, after a fast rise in theblood concentration of the drug, the intracellular free drugconcentration will rise slowly during influx at the front (at theside of the blood vessel). Because drug transport through theplasma membrane at the back of the cell (toward the neighbor-ing cell) depends on this intracellular free drug concentration, itwill also be slow and rise slowly (the latter is due to strongintracellular binding). In combination with a low paracellulartransport in epithelial cancer, the result may be a nonhomog-enous drug distribution in tissue. It must be noted that underthese conditions, the drug molecules are not hampered by slowconvective flows in the opposite direction driven by a pressuregradient.

In our search for gradientsin vivo, we have used com-puterized fluorescence microscopy, exploiting the autofluo-rescence of doxorubicin. Earlier reports on doxorubicin tissuedistribution in paraffin sections used either direct autofluo-rescence measurement by conventional fluorescence micros-copy (18) or indirect measurement using antibodies againstthe drug (19). The drug may be partly lost during samplepreparation with both methods. Both studies reported heter-ogeneous drug distribution throughout the tissue. To avoidsuch drug loss, we used cryosections for measuring doxoru-bicin fluorescence without further contact of the sectionswith solvents. In mice, Hoechst 33342 dye has been reportedto show steep gradients in biopsies alongside microvessels(20). Previously, doxorubicin gradients in tissue have beenobserved in mice, using an ovarian tumor model and i.p.injection (21). In our search for doxorubicin gradients afteri.v. injection of the drug in patients, we overlayed digitalpictures of doxorubicin fluorescence with immunohistochem-istry pictures of CD31 (staining of endothelial cells). Thisprocedure allowed us to determine the spatial distribution ofdoxorubicin in relation to microvessels.

Received 10/6/98; revised 3/15/99; accepted 3/19/99.The costs of publication of this article were defrayed in part by thepayment of page charges. This article must therefore be hereby markedadvertisementin accordance with 18 U.S.C. Section 1734 solely toindicate this fact.1 To whom requests for reprints should be addressed, at Department ofMedical Oncology, Room BR232, Free University Hospital, P. O. Box7057, 1007MB Amsterdam, the Netherlands. Phone: 31-20-444-2603;Fax: 31-20-444-3844; E-mail: j.lankelma@azvu.nl.

1703Vol. 5, 1703–1707, July 1999 Clinical Cancer Research

Research. on June 22, 2020. © 1999 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

1704Doxorubicin Gradients in Human Breast Cancer

Research. on June 22, 2020. © 1999 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

Patients with locally advanced breast cancer were treatedaccording to a moderately high-dose chemotherapy protocol inwhich doxorubicin was an important ingredient (22, 23).

MATERIALS AND METHODSChemotherapy. Chemotherapy consisted of moder-

ately high-dose doxorubicin (90 mg/m2 body surface) andhigh-dose cyclophosphamide (1000 mg/m2) on day 1, fol-lowed by granulocyte macrophage colony-stimulating factor(250 mg/m2) s.c. or i.v. on days 2–11. Cycles (total, sixcycles) were repeated every 3 weeks. Tumor biopsies ofapproximately 0.53 0.5 3 0.5 cm in size were taken underlocal or general anesthesia after oral consent and snap-frozenin liquid nitrogen. Cryosections of the whole biopsies werecut for imaging of doxorubicin fluorescence and subsequentimmunostaining.

Doxorubicin Imaging by CLSM. 2 A CLSM (type TCS4D; Leica Heidelberg, Heidelberg, Germany) was used. Flu-orescence images were made using the autofluorescence ofdoxorubicin. A krypton-argon laser line (488 nm) was usedfor excitation of doxorubicin, and a long pass filter (550 nm)was used for detection of the emitted light. Images at low andhigh resolution were obtained throughout the biopsy. We didnot observe any change in the doxorubicin distribution duringhandling of the samples, even after keeping the sample atroom temperature for 5 h. Integrated nuclear doxorubicinfluorescence was obtained using Leica Q500MC Quantimetsoftware (version V01.01; Leica Cambridge Ltd., Cambridge,United Kingdom).

Immunohistochemical Staining. The endothelial cellsof blood vessels were stained immunohistochemically, usingthe same frozen sections as for doxorubicin imaging, asfollows. The cryosections were fixed in acetone on ice or in4% paraformaldehyde followed by acetone on ice. They wereincubated for 1 h with mouse antihuman CD31 monoclonalantibody (clone JC/70A; DAKO A/S, Glostrup, Denmark),followed by a 30-min incubation with rabbit antimouse biotincomplex (DAKO). Peroxidase was blocked with 3% hydro-gen peroxide for 30 min before incubation with ABComplex/horseradish peroxidase (DAKO) for 30 min. A red color wasdeveloped in 10 –30 min with the 3-amino-9-ethylcarbarolekit (Vector Laboratories, Inc., Burlingame, CA), and nucleiwere counterstained with hematoxylin. Hematoxylin imageswere obtained by combining red, green, and blue transmis-sion images using the same optics. Images at low and highresolution were again obtained throughout the biopsy with

CLSM. Overlaying of pictures (doxorubicin and immunoim-ages) was accomplished using Photo-Paint software (version6.00; 1995; Corel Co., Ottawa, Canada).

Materials. Doxorubicin HCl was obtained from Monte-dison Nederland (Rotterdam, the Netherlands). DNase I waspurchased from Boehringer Mannheim (Mannheim, Germany).

RESULTSDoxorubicin Distribution in Breast Cancer. In the

cryosections, doxorubicin was seen as a nuclear fluorescenceclearly distinguishable from background fluorescence (Fig. 1),which was predominantly from the cytoplasm (Fig. 1A). Foreach patient (Fig. 1,B–J), both doxorubicin distribution patternsand CD31 immunohistochemical staining of the same area ofthe same section are represented. Doxorubicin gradients wereobserved most clearly in the ductal invasive carcinomas show-ing tumor islets in the biopsies, which were taken at 2 h afterinjection (Fig. 1,D–F). These doxorubicin gradients were notdetected in the connective tissue. Also, no clear gradients wereobserved in patients with invasive lobular cancer with moreconnective tissue and strands of tumor cells (indian files; Fig. 1,B and C). Occasionally, connective tissue showed bands of

2 The abbreviation used is: CLSM, confocal laser scanning microscope/microscopy.

Fig. 2 Mean nuclear doxorubicin fluorescence (averaged pixel inten-sity; expressed in arbitrary units)versusdistance to the border (with theconnective tissue) of the tumor islet of a representative part of thecorresponding images in Fig. 1.A corresponds to Fig. 1D;B correspondsto Fig. 1E; C corresponds to Fig. 1H; D corresponds to Fig. 1I.

Fig. 1 Doxorubicin (nuclear) fluorescence in cryosections of locally advanced breast cancer biopsies att 5 0 (A), 2 h (B–F), and 24 h (G–J) afterthe first i.v. injection of doxorubicin (dose, 90 mg/m2 body surface). A glow look-up table was used, with the lightest colors indicating the highestdoxorubicin concentrations. CD31 immunohistochemical staining combined with hematoxylin counterstaining of the very same biopsy area is showndirectly to theright of the corresponding doxorubicin fluorescence image. Each doxorubicin image is from a different patient, except forA (control)andD, which are from the same patient.

1705Clinical Cancer Research

Research. on June 22, 2020. © 1999 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

fluorescence (Fig. 1,D, F, andG); however, this fluorescencecould be distinguished morphologically from typical nucleardoxorubicin fluorescence.

Gradient Steepness. In four patients, the average nu-clear doxorubicin fluorescence was plotted against the distancefrom the nearby rim of the islet to demonstrate the steepness ofthe gradient (Fig. 2). Less steep gradients were found 24 h afterinjection, as compared to gradients at 2 h after injection.

Distribution at Lower Magnification. Fig. 3 shows anoverview over a larger part of one of the sections in pseudocolors(red, doxorubicin; green, CD31) with the highest doxorubicinconcentrations around the microvessels (103 magnification) andgradients extending from the capillaries (403 magnification).

Previously, doxorubicin concentrations (averaged perweight of tissue) in biopsies have been measured by high-performance liquid chromatography (24, 25). No metabolitescould be detected in breast cancer biopsies (25), indicating thatthe nuclear fluorescence comes only from doxorubicin itself.CD31 immunohistochemical staining indicated the presence ofcapillary vessels in the connective tissue. To compare the datafrom tumor with structurally similar normal tissue, skin biopsieswere taken from three patients after doxorubicin administration.Doxorubicin gradients were found in all three of the biopsiesexamined (data not shown).

DISCUSSIONGenerally, small molecules withMr , 1000 show a more

rapid diffusion into tissues than large molecules, such as antibodiesor viral vectors. Under conditions of extensive intracellular bindingand a low plasma membrane transport rate, penetration in the tumorislets of such drugs, such as doxorubicin (Mr 543.54), can beslowed down considerably. Here we show the slow penetration ofthe small-molecule drug doxorubicin in patients, resulting in gra-dients in clinical biopsies of solid tumors. Cells in the center of the

tumor islets, which are the most remote from the microvessels, areexposed to lower drug concentrations in the surrounding extracel-lular fluid compared to the cells in the periphery of the tumor islets.In general, the three-dimensional gradient toward the center of anislet may be steeper than the observed two-dimensional image, dueto the fact that the cross-sections of the islets generally will not runthrough the center of the islets. In that case, the distance from a cellinside such an islet to the islet boundary will be shorter than thatobserved in the two-dimensional cross-sections; in reality, thedoxorubicin concentration gradient will be even steeper than thatmeasured. It could be argued that when the gradient reverses duringdrug clearance from the blood, the drug concentration in cells at theperiphery might become lower when compared to cells at the centerof the islet. However, back supply from the inner layers will bufferthis fall in concentration. Diffusion will be slow through both thetranscellular and the paracellular routes: it will be slow through theparacellular route because of the small volume of the intercellularspace; and it will be slow through transcellular route because ofslow membrane passage and high intracellular binding (mainly toDNA). To understand the generation of doxorubicin gradients,further development of transport models determining the relativecontributions of factors affecting the total transport will be neces-sary.

As an alternative scheme for doxorubicin administrationper bolus injection, a continuous 96-h infusion has been appliedto patients for the treatment of metastatic breast cancer andsarcoma. This method of administration had a favorable effecton life-threatening heart toxicity (26, 27). The comparable clin-ical antitumor efficacy at the relatively low plasma concentra-tions during infusion, when compared to the high concentrationsduring the first few hours after i.v. injection, could partly be dueto a more homogenous doxorubicin tissue distribution in cancerislets. The kinetics of drug in the blood shows that after i.v.injection, the doxorubicin plasma concentration declined from

Fig. 3 Doxorubicin distribution (red) at 2 h after doxorubicin injection of patient K. CD31 immunostaining was visualized with a secondFITC-labeled antibody (green). Superimposition of both colors yieldsyellow. The highest nuclear doxorubicin concentrations were observed nearmicrovessels. Magnifications:310 (A); 340 (B andC).

1706Doxorubicin Gradients in Human Breast Cancer

Research. on June 22, 2020. © 1999 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

15 to 3mg/liter in the 6–96-h period after injection. During thefirst few hours after injection, the concentration declines muchmore rapidly from a level 2 orders of magnitude higher (24).With low effective drug diffusion into the tumor islet, thisprofile leads to the development of a gradient, especially in thefirst few hours, when the outer tumor cells of an islet areexposed to relatively high concentrations. Later,e.g.,at 24 h, thetissue gradient is less steep due to diffusion and the decline ofthe plasma concentration. During continuous infusion at a com-parable total dose, the plasma concentration gradually rose to 15mg/liter at 24 h after the start of the infusion (28). In that case,the high initial concentrations after i.v. injection are missing;therefore, the tumor islets will experience a much more steadyconcentration resulting in gradients that are less steep.

Doxorubicin-based chemotherapy of breast cancer fol-lowed by surgery (22, 23) can be an effective treatment forlocally advanced disease. To avoid possible memory effects ofprevious cycles, we examined biopsies after the first doxorubi-cin injection. This study suggests that during chemotherapycycles, the doxorubicin will result in a peeling off of cells fromtumor islets. Of the 10 patients shown, 3 patients (Fig. 1,B, C,andH) had no residual tumor cells in their breast after six cyclesof chemotherapy. Two of these patients (Fig. 1,B and C)showed no doxorubicin gradients in their breast carcinoma. Thisobservation underscores tissue transport of a drug as an impor-tant component of drug resistance. This study warrants addi-tional studies on the importance of tumor islets of breast cancercells as a negative prognostic marker in cancer chemotherapy.

ACKNOWLEDGMENTSWe thank Aafke Honkoop for obtaining the patient samples, and

we are most indebted to the patients who were willing to make tissueavailable without personal benefit. We thank Dr. Andy Ryan for gram-matical correction of the manuscript.

REFERENCES1. Ghadially, F. N. Cell junctions.In: Ultrastructural Pathology of theCell and Matrix, 2nd ed., pp. 797–820. London: Butterworths, 1982.2. Boehm, T., Folkman, J., Browder, T., and Oreilly, M. S. Antiangio-genic therapy of experimental cancer does not induce acquired drugresistance. Nature (Lond.),390: 404–407, 1997.3. Adson, A., Raub, T. J., Burton, P. S., Barsuhn, C. L., Hilgers, A. R.,Audus, K. L., and Ho, N. F. Quantitative approaches to delineateparacellular diffusion in cultured epithelial cell monolayers. J. Pharm.Sci., 83: 1529–1536, 1994.4. Sutherland, R. M. Cell and environment interactions in tumor mi-croregions: the multicell spheroid model. Science (Washington DC),240: 177–184, 1988.5. Durand, R. E. Flow cytometry studies of intracellular Adriamycin inmulticell spheroidsin vitro. Cancer Res.,41: 3495–3498, 1981.6. Gottesman, M. M., and Pastan, I. Biochemistry of multidrug resist-ance mediated by the multidrug transporter. Annu. Rev. Biochem.,62:385–427, 1993.7. Chaudhary, P. M., and Roninson, I. B. Expression and activity ofP-glycoprotein, a multidrug efflux pump, in human hematopoietic stemcells. Cell,66: 85–94, 1991.8. Bradley, G., Juranka, P. F., and Ling, V. Mechanism of multidrugresistance. Biochim. Biophys. Acta,948: 87–128, 1988.9. Cole, S. P., Bhardwaj, G., Gerlach, J. H., Mackie, J. E., Grant, C. E.,Almquist, K. C., Stewart, A. J., Kurz, E. U., Duncan, A. M., and Deeley,R. G. Overexpression of a transporter gene in a multidrug-resistant humanlung cancer cell line. Science (Washington DC),258: 1650–1654, 1992.

10. Kool, M., de Haas, M., Scheffer, G. L., Scheper, R. J., van Eijk, J.,Juijn, J. A., Baas, F., and Borst, P. Analysis of expression of cMOAT(MRP2), MRP3, MRP4, and MRP5, homologues of the multidrugresistance-associated protein gene (MRP1), in human cancer cell lines.Cancer Res.,57: 3537–3547, 1997.

11. Lankelma, J., Spoelstra, E. C., Dekker, H., and Broxterman, H. J.Evidence for daunomycin efflux from multidrug-resistant 2780AD hu-man ovarian carcinoma cells against a concentration gradient. Biochim.Biophys. Acta,1055: 217–222, 1990.

12. Broxterman, H. J., Giaccone, G., and Lankelma, J. Multidrug re-sistance proteins and other drug transport-related resistance to naturalproduct agents. Curr. Opin. Oncol.,7: 532–540, 1995.

13. Broxterman, H. J., Kuiper, C. M., Schuurhuis, G. J., Tsuruo, T.,Pinedo, H. M., and Lankelma, J. Increase of daunorubicin and vincris-tine accumulation in multidrug resistant human ovarian carcinoma cellsby a monoclonal antibody reacting with P-glycoprotein. Biochem. Phar-macol.,37: 2389–2393, 1988.

14. Spoelstra, E. C., Westerhoff, H. V., Dekker, H., and Lankelma, J.Kinetics of daunorubicin transport by P-glycoprotein of intact cancercells. Eur. J. Biochem.,207: 567–579, 1992.

15. Jain, R. K. Delivery of molecular medicine to solid tumors. Science(Washington DC),271: 1079–1080, 1996.

16. Jain, R. K. Delivery of novel therapeutic agents in tumors: physi-ological barriers and strategies. J. Natl. Cancer Inst.,81: 570–576, 1989.

17. Morrison, P. F., Bungay, P. M., Hsiao, J. K., Ball, B. A., Mefford,I. N., and Dedrick, R. L. Quantitative microdialysis: analysis of tran-sients and application to pharmacokinetics in brain. J. Neurochem.,57:103–119, 1991.

18. Stallard, S., Morrison, J. G., George, W. D., and Kaye, S. B.Distribution of doxorubicin to normal breast and tumour tissue inpatients undergoing mastectomy. Cancer Chemother. Pharmacol.,25:286–290, 1990.

19. Henneberry, H. P., and Aherne, G. W. Visualisation of doxorubicinin human and animal tissues and in cell cultures by immunogold-silverstaining. Br. J. Cancer,65: 82–86, 1992.20. Smith, K. A., Hill, S. A., Begg, A. C., and Denekamp, J. Validationof the fluorescent dye Hoechst 33342 as a vascular space marker intissues. Br. J. Cancer,57: 247–253, 1988.21. Ozols, R. F., Locker, G. Y., Doroshow, J. H., Grotzinger, K. R., Myers,C. E., and Young, R. C. Pharmacokinetics of Adriamycin and tissuepenetration in murine ovarian cancer. Cancer Res.,39: 3209–3214, 1979.22. Hoekman, K., Wagstaff, J., van Groeningen, C. J., Vermorken, J. B.,Boven, E., and Pinedo, H. M. Effects of recombinant human granulocyte-macrophage colony-stimulating factor on myelosuppression induced bymultiple cycles of high-dose chemotherapy in patients with advanced breastcancer. J. Natl. Cancer Inst.,83: 1546–1553, 1991.23. Honkoop, A. H., Hoekman, K., Wagstaff, J., van Groeningen, C. J.,Vermorken, J. B., Boven, E., and Pinedo, H. M. Continuous infusion orsubcutaneous injection of granulocyte-macrophage colony-stimulatingfactor: increased efficacy and reduced toxicity when given subcutane-ously. Br. J. Cancer,74: 1132–1136, 1996.24. Rossi, C., Gasparini, G., Canobbio, L., Galligioni, E., Volpe, R.,Candiani, E., Toffoli, G., and d’Incalci, M. Doxorubicin distribution inhuman breast cancer. Cancer Treat. Rep.,71: 1221–1226, 1987.25. Cummings, J., and McArdle, C. S. Studies on thein vivodispositionof Adriamycin in human tumours which exhibit different responses tothe drug. Br. J. Cancer,53: 835–838, 1986.26. Weiss, R. B. The anthracyclines: will we ever find a better doxo-rubicin? Semin. Oncol.,19: 670–686, 1992.27. Legha, S. S., Benjamin, R. S., Mackay, B., Yap, H. Y., Wallace, S.,Ewer, M., Blumenschein, G. R., and Freireich, E. J. Adriamycin therapyby continuous intravenous infusion in patients with metastatic breastcancer. Cancer (Phila.),49: 1762–1766, 1982.28. Speth, P. A. J., Linssen, P. C. M., Holdrinet, R. S. G., and HaanenC. Plasma and cellular Adriamycin concentrations in patients withmyeloma treated with ninety-six-hour continuous infusion. Clin. Phar-macol. Ther.,41: 661–665, 1987.

1707Clinical Cancer Research

Research. on June 22, 2020. © 1999 American Association for Cancerclincancerres.aacrjournals.org Downloaded from

1999;5:1703-1707. Clin Cancer Res   Jan Lankelma, Henk Dekker, Rafael Fernández Luque, et al.   Doxorubicin Gradients in Human Breast Cancer

  Updated version

  http://clincancerres.aacrjournals.org/content/5/7/1703

Access the most recent version of this article at:

   

   

  Cited articles

  http://clincancerres.aacrjournals.org/content/5/7/1703.full#ref-list-1

This article cites 27 articles, 5 of which you can access for free at:

  Citing articles

  http://clincancerres.aacrjournals.org/content/5/7/1703.full#related-urls

This article has been cited by 20 HighWire-hosted articles. Access the articles at:

   

  E-mail alerts related to this article or journal.Sign up to receive free email-alerts

  Subscriptions

Reprints and

  .pubs@aacr.orgDepartment at

To order reprints of this article or to subscribe to the journal, contact the AACR Publications

  Permissions

  Rightslink site. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC)

.http://clincancerres.aacrjournals.org/content/5/7/1703To request permission to re-use all or part of this article, use this link

Research. on June 22, 2020. © 1999 American Association for Cancerclincancerres.aacrjournals.org Downloaded from