Purpose: The high molecular weight and binding affinity of trastuzumab, a monoclonal antibody in use for treatment of breast cancers overexpressing human epidermal growth factor receptor type 2 (HER2), in combination with microenvironmental factors, may limit its distribution and efficacy. We assessed and mapped the distribution of systemically given, unlabeled trastuzumab at micrometer resolution in tumor xenografts using immunohistochemistry.

Experimental Design: Mice bearing MDA-435/LCC6HER2 xenografts were given single doses of 4 or 20 mg/kg unlabeled trastuzumab with tumor harvest at various time points thereafter; bound trastuzumab was imaged directly in tumor cryosections using fluorescently tagged antihuman secondary antibodies. Combinations of additional markers, including HER2, 5-bromo-2-deoxyuridine, CD31, DioC7(3), desmin, and collagen IV were also mapped on the same tumor sections.

Results: Distribution of trastuzumab in MDA-435/LCC6HER2 tumors is found to be heterogeneous, with tumor margins saturating more thoroughly in doses and times analyzed. Considerable intervessel heterogeneity is also seen. For example, in unsaturated tissues, there remain perfused vessels without any trastuzumab in addition to vessels with a few layers of positively stained perivascular cells, in addition to vessels with bound drug up to 150 μm away. This heterogeneity is independent of HER2 expression, microvessel density, and perfusion. A slightly greater proportion of vessels were associated with pericytes in sections with greater trastuzumab saturation, but this would not adequately account for observed heterogeneous trastuzumab distribution.

Conclusions: Complete penetration of trastuzumab in tumor tissue was not seen in our study, leaving the possibility that inadequate distribution may represent a mechanism for resistance to trastuzumab.

Many anticancer agents may have limited efficacy due to their inability to distribute in tumor tissue and expose all cells to therapeutic doses (1). Tumor cells exhibit uneven responses to anticancer therapy due to their proximity to vasculature and microenvironment where factors such as oxygen and nutrient availability influence drug sensitivity. Tumor vascular abnormalities that contribute to inefficient drug penetration include large and variable intervascular distances, complex and tortuous branching patterns, lack of lymphatic vessel drainage, and irregular vascular phenotypes related to maturity, angiogenic capacity, size, perfusion, and permeability (for a review, see ref. 2). These contributing abnormalities create regions of irregular or intermittent blood flow, slowed interstitial fluid velocity, and often elevated interstitial fluid pressure (3). The extravascular compartment of tumors is also highly variable and abnormal, with tumor-specific extracellular matrix density, characteristics and expression patterns (46), presence or absence of stromal tissues, such as inflammatory cells (7), and altered tumor cell expression of antigen, receptors (8), or structures for transport of drug into or out of tumor cells (9).

With this veritable gauntlet of factors capable of inhibiting access of drugs to their target tumor cells, in addition to pharmacokinetic variable such as consumption and metabolism of diffusing drugs, it is not surprising that many small molecule cancer therapeutics have shown evidence of poor tumor tissue penetration in both in vitro and in vivo model systems, including doxorubicin (10), gemcitabine (11), docetaxel, and paclitaxel (ref. 12; for a recent review, see ref. 1). Monoclonal antibodies can face added difficulties due to their high molecular weights, target specificity, kinetics of metabolism and internalization, as well as the patterns of antigen expression in combination with target-binding affinity (13, 14).

The human epidermal growth factor receptor type 2 (HER2) is overexpressed in 20% to 30% of all invasive breast carcinomas and is a prognostic factor of an aggressive tumor phenotype, increased relapse, and decreased overall survival (15). HER2 status is determined in patient tissue samples by pathologic analysis using fluorescence in situ hybridization and immunohistochemistry; patients classified as positive for HER2-overexpressing breast cancer are candidates for, and have been shown to benefit from, therapy that includes a humanized monoclonal antibody against HER2, trastuzumab (Roche; refs. 1619).

Several methods for investigating extravasation and tissue penetration of HER2-targeting antibodies and antibody fragments have been previously reported, including the use of fluorescent or radiolabeled drug preparations for imaging of tumor sections (20), intravital scintigraphic imaging (21), single-photon emission computed tomography (22), positron emission tomography (23), and magnetic resonance imaging (24). Additionally, the in vivo pharmacokinetics of trastuzumab have been analyzed at the single-molecule level using trastuzumab-conjugated fluorescent nanoparticles, called Q-dots, along with intravital imaging (25). Such studies can provide practical insight regarding strategies for the design of new immunotherapeutics and for the optimization of dosing and delivery regimes to result in maximal tissue penetration.

In this study, we show a quantitative approach to the study of trastuzumab penetration in tumors by directly imaging systemically administered, unlabeled trastuzumab in xenograft tumors using fluorescently tagged anti-human secondary antibodies on tissue cryosections. The microregional location of bound trastuzumab is then mapped in relation to several microenvironmental features, including vasculature, indicators of microvessel structural and functional integrity, HER2 expression, and cellular proliferation. This is the first study describing distribution of unlabeled trastuzumab in the context of the tumor microenvironment at micrometer resolution.

Reagents. Trastuzumab (2.2 mg/mL; Hoffman-La Roche) was provided by the British Columbia Cancer Agency pharmacy; dilutions were prepared in sterile 0.9% NaCl before injection. 5-Bromo-2-deoxyuridine (BrdUrd; Sigma Chemical), 30 mg/mL solution in 0.9% NaCl, was used to label S-phase cells. The fluorescent dye DioC7(3) (Molecular Probes), 0.6 mg/mL dissolved in 75% (v/v) DMSO:25% sterile H2O, was given i.v. as a marker of vessel perfusion (26).

Cells and cell culture. MDA-435/LCC6 cells (LCC6) and LCC6 cells stably transfected by electroporation with the mammalian expression vector pREP9 (Invitrogen) containing the 4.3-kb full-length human normal c-erbB2 cDNA to yield LCC6HER2 or with the empty vector pREP9 to yield LCC6Vector cells were kindly provided by Dr. Wieslawa Dragowska (British Columbia Cancer Research Center). Cells were maintained as monolayers using MEM/EBSS (HyClone) supplemented with 10% bovine growth serum (HyClone) and passaged every 3 to 4 days. Cells were selected using media containing 500 μg/mL G418. Cellular HER2 overexpression was ascertained by flow cytometry, as previously described (27).

Mice and tumors. Female, nonobese, diabetic, severe combined immunodeficient NOD-SCID mice were bred and maintained in our institutional specific pathogen–free animal facility in accordance with the Canadian Council on Animal Care guidelines. The experiments described in this manuscript were approved by the Animal Care Committee of the University of British Columbia. Mice were allowed free access to standard laboratory rodent food and water and were used between 8 and 24 weeks of age, weighing between 20 and 28 g. LCC6HER2 and LCC6Vector single cell suspensions (50 μL of 1.6 × 108 cells/mL) were implanted s.c. into the sacral region of mice. Mice were randomly assigned to experimental groups when their tumors reached 100 to 150 mm3 as measured using calipers of three orthogonal diameters (a, b, and c) using the formula volume: π/6(abc). Trastuzumab was administered via i.p. injection. Two hours before tumor excision, all mice were administered an S-phase marker at 1,500 mg/kg BrdUrd i.p.; 5 min before euthanasia and tissue excision, all mice were administered an i.v. perfusion marker of 75 μL 0.6 mg/mL DioC7(3).

Immunohistochemistry. Immediately after excision, tumors were frozen on an aluminum block held at −20°C, embedded in optimum cutting temperature (Tissue-TEK), and stored at −20°C until sectioning. All transverse sections were taken perpendicular to the cranial-caudal axis; serial step cryosections were obtained at 1-mm intervals. The general immunohistochemical procedure used has been previously reported (11). Briefly, 10-μm tumor cryosections were cut with a Cryostar HM560 (Microm International GmbH), air-dried, imaged for native DioC7(3) fluorescence, and fixed in 50% (v/v) acetone/methanol for 10 min at room temperature. All vasculature including nonperfused vessels, were identified using an antibody to PECAM/CD31 (BD PharMingen) with fluorescent Alexa 647 antirat secondary anti-body (Invitrogen). Trastuzumab was visualized using Alexa 546 antihuman secondary antibody (Invitrogen). Tissue sections were always imaged for trastuzumab before staining for HER2. HER2 not previously bound by trastuzumab was detected with a 1:10,000 dilution of 2.2 mg/mL trastuzumab and the same goat antihuman Alexa 546 secondary (Invitrogen) used for systemically given, bound trastuzumab detection. Pericytes were detected using a monoclonal antibody to desmin (clone D33, Signet antibodies) and fluorescent Alexa 488 antimouse secondary antibody (Invitrogen). Collagen IV was stained with a polyclonal antibody to collagen IV (Gene Tex) and antirabbit peroxidase-conjugated antibody (Sigma) developed with metal-enhanced 3,3′-diaminobenzidine substrate (Pierce). Slides stained for incorporated BrdUrd were treated with 4 mol/L HCl and neutralized in 0.1 mol/L sodium borate. Immunohistochemical detection used a monoclonal mouse anti-BrdUrd (clone BU33; Sigma) followed by an antimouse peroxidase-conjugated antibody (Sigma) and a metal-enhanced 3,3′-diaminobenzidine substrate (Pierce). Slides were then counterstained with hematoxylin, dehydrated, and mounted using Permount (Fisher Scientific) before imaging.

Image acquisition and overlay. The imaging system consisted of two set-ups (a and b): a robotic fluorescence microscope (axio Imager Z1a or IIIRS; Zeissb), monochrome CCD camera (Retiga 4000R, Qimaginga; or model 4922, Cohub), and a motorized slide loader with x-y stage (Ludl Electronic Productsa), or a custom-built motorized x-y stageb. Customized ImageJa or NIH-Image softwareb (public domain program developed at the NIH1

) running on G5a or G3b Macintosh computers (Apple Computers) allowed for automated tiling of adjacent microscope fields. Images of entire tumor cryosections up to 1 cm2 were captured at a resolution of 0.75 μma or 1.0 μmb per pixel. Grayscale images of trastuzumab and HER2 were overlaid using Adobe Photoshop CS (version 8.0); images of trastuzumab (green channel) and HER2 (red channel) were then combined using the darken mask, resulting in overlapped areas displayed as black. Grayscale CD31 and DioC7(3) images were thresholded, and a composite color image was produced (CD31 in dark blue prioritized over DioC7(3) in cyan) and overlaid on the images of trastuzumab (red) or combination images of HER2 (red) and trastuzumab (black).

Image analysis. Using the NIH-Image software application and user-supplied algorithms, fluorescent images were inverted and combinations of images of DioC7(3), CD31, D33, Coll IV, trastuzumab, HER2, BrdUrd, and hematoxylin staining from each tumor section aligned and cropped to tumor tissue boundaries and staining artifacts were removed. Percentage of positive staining was obtained using the proportion of pixels at intensities meeting or exceeding a threshold value determined to be above background. For distribution analysis relative to vasculature, each pixel in an image was sorted based on its distance relative to CD31-positive vasculature, and the average tissue intensity, in 1.5-μm increments from vasculature, was determined for each tumor. Data is displayed as a combined average of a minimum of four tumors. Data was normalized by subtracting average intensities of control images at each distance. For dual positive-staining analysis of CD31 in combination with additional markers, thresholds were set to identify staining above background, and a minimum of 20% overlap was required to classify CD31 objects as dual labeled. Necrotic tissue was cropped out from hematoxylin-stained sections for all analysis listed above; the proportion of necrotic tissue was obtained by dividing the total number of pixels in images after cropping by the total number of pixels in whole-tumor areas. Where appropriate, charts display means ± SE.

The distribution of trastuzumab within HER2 overexpressing xenografts was observed by examining tumors excised at a range of times after i.p. administration. Tumor cryosections were imaged and stained immunohistochemically for combinations of trastuzumab, HER2, endothelial cells, pericytes, collagen type IV, perfusion marker DioC7(3), and S-phase cells labeled with BrdUrd. Images of up to five markers were obtained from each tissue section.

Trastuzumab imaged directly in HER2 overexpressing xenografts in combination with immunohistochemical staining for HER2. Cryosections of MDA-435/LCC6HER2 xenografts from mice administered 20 mg/kg trastuzumab 3 h before tumor recovery (Fig. 1A and B) are compared with untreated tumors (Fig. 1C and D). Sections were imaged for the i.v. fluorescent dye, DioC7(3), (cyan) before immunostaining for CD31 (dark blue) and trastuzumab (black), which can be seen bound in the treated tumor at limited distances surrounding vasculature; note considerable intervessel heterogeneity with respect to the amount of associated trastuzumab (Fig. 1A). The same sections were subsequently stained for HER2, which is found to be present homogeneously throughout the field (Fig. 1B and D). Immunohistochemical staining for HER2 was done on tumors from all experiments to confirm relatively uniform expression (data not shown).

Fig. 1.

Staining for bound trastuzumab and HER2. Tumor cryosections are shown from MDA-435-LCC6HER2-overexpressing xenografts treated with 20 mg/kg trastuzumab for 3 h or left untreated. Overlaid images of a treated tumor show bound trastuzumab (black) relative to blood vessels (CD31; dark blue) and the perfusion marker DioC7(3) (cyan; A); additional staining of the same section for HER2 (red) shows that areas with no bound trastuzumab are overexpressing HER2 (B). An untreated MDA-435-LCC6HER2 tumor shows no bound trastuzumab (C) but relatively homogeneous HER2 expression (D). Similarly stained MDA-435-LCC6vector tumors display no bound trastuzumab or unbound HER2 in treated or untreated tumors (not shown).

Fig. 1.

Staining for bound trastuzumab and HER2. Tumor cryosections are shown from MDA-435-LCC6HER2-overexpressing xenografts treated with 20 mg/kg trastuzumab for 3 h or left untreated. Overlaid images of a treated tumor show bound trastuzumab (black) relative to blood vessels (CD31; dark blue) and the perfusion marker DioC7(3) (cyan; A); additional staining of the same section for HER2 (red) shows that areas with no bound trastuzumab are overexpressing HER2 (B). An untreated MDA-435-LCC6HER2 tumor shows no bound trastuzumab (C) but relatively homogeneous HER2 expression (D). Similarly stained MDA-435-LCC6vector tumors display no bound trastuzumab or unbound HER2 in treated or untreated tumors (not shown).

Close modal

Distribution of trastuzumab over time. Mice bearing MDA-435/LCC6HER2 xenografts were given single i.p. doses of 4 mg/kg trastuzumab. Representative high resolution composite images for whole tumor sections obtained 2 to 2.5 mm from the tumor edge were generated as described above, with a magnified portion of an untreated control tumor shown in Fig. 2A. Similar images of treated tumors show trastuzumab staining (red) proximal to vasculature (CD31 in dark blue, perfusion marker DioC7(3) in cyan) at 8 h (Fig. 2B), well distributed by 24 h (Fig. 2C), decreased in overall levels at 72 h (Fig. 2D), and barely detectable above control levels by 120 h (Fig. 2E). Quantitative analysis was done on whole tumor sections via selection of a threshold for bound trastuzumab staining; the extracellular binding pattern of this marker does not saturate all pixels representative of a cell, and therefore, the 15% positive tissue is not a reflection of the proportion of cells positive for bound trastuzumab but is instead a relative indication of amount of bound drug. Results indicate that after a 4 mg/kg dose, detectable levels are found as early as 3 h, with high levels of bound trastuzumab at 8 to 24 h compared with 72 to 120 h, wherein levels are reduced but are still above background (n = 5 per group; Fig. 2F).

Fig. 2.

Distribution of trastuzumab over time. The relative amount of bound trastuzumab in cryosections obtained 2 to 2.5 mm from the tumor edge of MDA-435-LCC6HER2 tumors is assessed using quantitative analysis of immunohistochemical staining. Representative portions of whole-tumor sections are shown with trastuzumab (red), CD31 (dark blue), and perfusion dye, DioC7(3) (cyan): an untreated control tumor (A), a single dose of 4 mg/kg trastuzumab at 8 h (B), 24 h (C), 72 h (D), and 96 h (E). The average fraction of tissue stained positively for bound trastuzumab shows detectable levels as early as 3 h, peak levels at 24 h, with significant decreases at 72 and 120 h (F).

Fig. 2.

Distribution of trastuzumab over time. The relative amount of bound trastuzumab in cryosections obtained 2 to 2.5 mm from the tumor edge of MDA-435-LCC6HER2 tumors is assessed using quantitative analysis of immunohistochemical staining. Representative portions of whole-tumor sections are shown with trastuzumab (red), CD31 (dark blue), and perfusion dye, DioC7(3) (cyan): an untreated control tumor (A), a single dose of 4 mg/kg trastuzumab at 8 h (B), 24 h (C), 72 h (D), and 96 h (E). The average fraction of tissue stained positively for bound trastuzumab shows detectable levels as early as 3 h, peak levels at 24 h, with significant decreases at 72 and 120 h (F).

Close modal

Heterogeneous distribution of trastuzumab through whole tumors. Mice bearing MDA-435/LCC6HER2 xenografts were given a single dose of 20 mg/kg trastuzumab with tumor excision at 3 or 26 h. Transverse serial step cryosections (1-mm interval) were obtained for each tumor and were stained as described above. A greater proportion of tumor tissue positive for bound trastuzumab was found in marginal sections (1 mm from the tumor edge) compared with those from more central tumor sections (4 mm from the tumor edge) in both the 3 and 26 h time points, as shown in images of whole-tumor sections (Fig. 3A and B). Quantitative analysis of the percentage of pixels stained positively for trastuzumab above threshold is expressed as a function of distance from the tumor edge (Fig. 3C, left). Additional analysis determined the average intensity of trastuzumab staining as a function of distance from nearest microvessel; data shown are averages for a minimum of four tumors and are normalized by subtracting average untreated control intensities at each distance. After treatment, the average intensity throughout the tumor cord, up to 150 μm from vasculature, is lower in the sections 4 mm from the tumor edge compared with those only 1 mm from the edge in both the 3 and 26 h groups (Fig. 3C, center and right).

Fig. 3.

Heterogeneous distribution of trastuzumab through whole tumors. Tumors grown to 8 to 10 mm in diameter were dosed with 20 mg/kg trastuzumab and harvested at 3 h (A) or 26 h (B); 10 μm thick transverse sections were obtained at 1mm intervals from the tumor edge, were imaged for perfusion marker DioC7(3) (cyan), and stained for bound trastuzumab (red) and CD31 (dark blue). Examples of vessels with very little or no extravasating trastuzumab are illustrated (green arrows). Those sections obtained closer to the gross tumor edge of both the 3 h (A, left) and 26 h (B, left) show a greater proportion of tissue stained for trastuzumab than the more heterogeneous central sections (A and B, right). Quantitative analysis shows percentage of pixels positive for trastuzumab in sections at increasing distances from the tumor edge (C, left). The microregional distribution of trastuzumab as a function of distance from vasculature was analyzed on sections obtained 1 mm from tumor edge and compared with those at 4 mm; data is shown for tumors harvested at 3 h (C, middle) and 26 h (C, right) after treatment.

Fig. 3.

Heterogeneous distribution of trastuzumab through whole tumors. Tumors grown to 8 to 10 mm in diameter were dosed with 20 mg/kg trastuzumab and harvested at 3 h (A) or 26 h (B); 10 μm thick transverse sections were obtained at 1mm intervals from the tumor edge, were imaged for perfusion marker DioC7(3) (cyan), and stained for bound trastuzumab (red) and CD31 (dark blue). Examples of vessels with very little or no extravasating trastuzumab are illustrated (green arrows). Those sections obtained closer to the gross tumor edge of both the 3 h (A, left) and 26 h (B, left) show a greater proportion of tissue stained for trastuzumab than the more heterogeneous central sections (A and B, right). Quantitative analysis shows percentage of pixels positive for trastuzumab in sections at increasing distances from the tumor edge (C, left). The microregional distribution of trastuzumab as a function of distance from vasculature was analyzed on sections obtained 1 mm from tumor edge and compared with those at 4 mm; data is shown for tumors harvested at 3 h (C, middle) and 26 h (C, right) after treatment.

Close modal

Qualitative observations of individual sections and close examination of bound trastuzumab around individual vessels reveals considerable intervessel heterogeneity. For example, in tumors recovered just 3 h after trastuzumab administration, neighboring vessels that were perfused at the time of tumor recovery [CD31 objects also positive for DioC7(3)] or seem to be similar in size may show varying amounts of bound trastuzumab ranging from just a few cell layers to up to 50 μm (Fig. 4A) or may in fact show no bound drug at all (arrows). In tissues that have unsaturated trastuzumab binding even at later time points, there can remain vessels without any trastuzumab (Fig. 3B, arrows), in addition to vessels with a few layers of perivascular cells staining positive for bound drug (Figs. 3B and 5B). Areas remaining unsaturated with trastuzumab were not consistently located in the center of tumor sections nor do they necessarily have low vascular density or lower rates of perfusion (Fig. 4B).

Fig. 4.

Heterogeneous microregional distribution of trastuzumab. Tumors grown to 8 to 10 mm in diameter were dosed with 20 mg/kg trastuzumab and harvested at 3 h (A) or 26 h (B); 10 μm cryosections were imaged for perfusion marker DioC7(3) (cyan) and stained for bound trastuzumab (red) and CD31 (dark blue). Trastuzumab is seen to extravasate in a heterogeneous pattern in all regions of the tumor sections at 3 h (A, right); inset (A, left) illustrates presence of many perfused vessels not associated with significant amounts of trastuzumab (green arrows). At 26 h, tissue free of bound trastuzumab can still be found on cryosections (B, middle) in areas demonstrating both a high density of perfused vessels (B, left) and low density (B, right). All scale bars, 200 μm.

Fig. 4.

Heterogeneous microregional distribution of trastuzumab. Tumors grown to 8 to 10 mm in diameter were dosed with 20 mg/kg trastuzumab and harvested at 3 h (A) or 26 h (B); 10 μm cryosections were imaged for perfusion marker DioC7(3) (cyan) and stained for bound trastuzumab (red) and CD31 (dark blue). Trastuzumab is seen to extravasate in a heterogeneous pattern in all regions of the tumor sections at 3 h (A, right); inset (A, left) illustrates presence of many perfused vessels not associated with significant amounts of trastuzumab (green arrows). At 26 h, tissue free of bound trastuzumab can still be found on cryosections (B, middle) in areas demonstrating both a high density of perfused vessels (B, left) and low density (B, right). All scale bars, 200 μm.

Close modal
Fig. 5.

Quantitative analysis of tumor microenvironment. Analyses of vascular architectural features were done on MDA-435-LCC6HER2 tumors treated with 20 mg/kg trastuzumab for either 3 or 26 h (A); transverse sections were obtained at 1 mm intervals from the tumor edge. No differences are seen in the density of CD31-stained microvessels (Ai), or the fraction of vessels staining dual positive for perfusion marker DioC7(3) (Aii), or collagen type IV (Aiii); however, dual staining of CD31 with pericyte marker desmin D33 shows a decreasing trend at distances farther from the tumor edge (Aiv). Tumor microenvironmental features were also assessed at longer time points after a single 4 mg/kg dose of trastuzumab (B), with no significant changes in the fraction of perfused vessels (Bi), S-phase cells detected using BrdUrd incorporation (Bii), or the relative fraction of necrotic tumor tissue at time points up to 96 h after administration (Biii); however, the fraction of necrotic tissue that stains positive for bound trastuzumab does increase with time after treatment (Biv).

Fig. 5.

Quantitative analysis of tumor microenvironment. Analyses of vascular architectural features were done on MDA-435-LCC6HER2 tumors treated with 20 mg/kg trastuzumab for either 3 or 26 h (A); transverse sections were obtained at 1 mm intervals from the tumor edge. No differences are seen in the density of CD31-stained microvessels (Ai), or the fraction of vessels staining dual positive for perfusion marker DioC7(3) (Aii), or collagen type IV (Aiii); however, dual staining of CD31 with pericyte marker desmin D33 shows a decreasing trend at distances farther from the tumor edge (Aiv). Tumor microenvironmental features were also assessed at longer time points after a single 4 mg/kg dose of trastuzumab (B), with no significant changes in the fraction of perfused vessels (Bi), S-phase cells detected using BrdUrd incorporation (Bii), or the relative fraction of necrotic tumor tissue at time points up to 96 h after administration (Biii); however, the fraction of necrotic tissue that stains positive for bound trastuzumab does increase with time after treatment (Biv).

Close modal

Vascular architectural features remain consistent throughout tumors with the exception of vascular association with pericytes. Tumor sections collected as above from mice treated with 20 mg/kg trastuzumab were stained for markers of vascular phenotype in serial step sections (1-mm interval) up to 4 mm from the tumor edge. No differences were found in microvascular density (Fig. 5A, i). Similarly, no difference was detected in the fraction of perfused vessels or vessels associated with presence of basal lamina, as detected by colocalization of CD31 and DioC7(3) or collagen IV, respectively (ii and iii). In contrast, a trend of increased association of D33 staining, a marker of pericytes, was found in association with CD31 in sections closest to the tumor edge (iv).

Tumor microenvironmental features do not significantly change after a single 4 mg/kg dose of trastuzumab. No significant microenvironmental changes were observed in MDA-435/LCC6HER2 tumors at time points up to 120 h after a single 4 mg/kg dose of trastuzumab. Quantitative image analysis of cryosections obtained 2 to 2.5 mm from the tumor edge showed no change with time in microvessel density, the fraction of perfused vessels (Fig. 5B, i) or the fraction of BrdUrd incorporating cells analyzed as both percentage of positive pixels (ii) or average intensity of staining as a function of distance from blood vessels (data not shown). Although the relative fraction of tissue identified as being necrotic by examination of hematoxylin-stained tissue did not change (iii), the fraction of necrotic tissue staining above threshold for bound trastuzumab does increase with time after treatment (iv).

This study presents the first direct visualization of systemically administered trastuzumab in xenograft tumor tissue using immunohistochemistry, an approach that could be useful to many studies investigating distribution of monoclonal antibody therapeutics. This simple technique enabled qualitative and quantitative analysis of trastuzumab distribution to be assessed at high resolution in the context of additional tumor microenvironmental features, including HER2 expression, vasculature, and markers of vascular phenotype.

The tumor model MDA-435/LCC6HER2 is derived from the MDA-MB-435 cell line, which was formerly thought to be a human breast carcinoma, but genetic profiling has recently revealed it to have melanoma origins (28). The intention of the present study was to look at trastuzumab penetration and distribution in an in vivo preclinical model of tumor tissue overexpressing the HER2 antigen.

Trastuzumab has previously been shown to have anticancer efficacy against this tumor line (27), although results presented here show no changes in microvessel density, perfusion, proportion of S-phase tumor cells, or necrotic tissue were found up to 120 h after a single 4 mg/kg dose (Fig. 5A). It was noted, however, that the necrotic regions of treated tumors (72-120 h) had significant levels of bound trastuzumab, but it is unclear if this is a relevant phenomenon or an experimental artifact. Despite limited penetration of trastuzumab at some blood vessels, the majority of tissue is exposed to trastuzumab by 24 h, suggesting that if a measurable effect on tumor cell proliferation were to occur in this model under these conditions, it would be detectable in gross tissue analyses. However, this panel of markers assessing the effects of trastuzumab is not comprehensive and would not necessarily be expected to reflect its activity at such short time points after a single dose. The multimarker, serially immunostained model enabled simplified analysis of multiple markers per tumor section; future studies examining the effects of trastuzumab over prolonged periods with multiple dosage regimens may be able to use a similar system with these or additional pharmacodynamic markers to investigate mechanisms of activity of trastuzumab and relate them to the distribution and retention of bound drug or HER2 expression. Specific immunohistochemical techniques developed by Glazyrin et. al. (19) use trastuzumab as a primary antibody for the detection of HER2 status in untreated clinical samples despite the presence of additional human antibodies; it is possible that these same techniques could be extended to examine the distribution of bound trastuzumab, which could provide highly valuable data regarding the ability of trastuzumab to distribute and accumulate in treated patients.

Accumulation of trastuzumab in tumor xenografts has previously been shown to reach peak concentrations within a few days of administration (20, 23, 29, 30) and substantial accumulation in the interstitial space has been detected at 2 h after administration (25). In this model, results are comparable with detectable levels as early as 3 h, significant levels at 8 to 24 h, and reduction by 72 h after a 4 mg/kg administration (Fig. 2). This dose was selected due to its use in previous studies with HER2-overexpressing xenograft tumors, as well as its similarity to clinically relevant doses (31, 32). These results suggest that the model used, MDA-435/LCC6HER2, exhibits trastuzumab binding and metabolism characteristics similar to those seen in other models.

Microregional distribution of trastuzumab was found to be heterogeneous, with higher levels of bound drug found in transverse tissue sections obtained proximal to the tumor edge (1 mm) compared with more central sections (4 mm; Fig. 3). This heterogeneity is not explained by variable expression of HER2, which was found to be expressed in tumor tissues not bound by systemically administered trastuzumab (Fig. 1). Considerable intervessel heterogeneity was noted, wherein varying rates of extravasation and extravascular distribution of trastuzumab from tumor vessels were often seen (Figs. 4 and 5) in all areas of the tumor, often with neighboring vessels exhibiting completely different characteristics. Quantitative data was obtained to investigate whether vascular phenotype may be partially responsible for the greater degree of trastuzumab tissue saturation found in sections proximal to the tumor edge. A straightforward explanation for decreased trastuzumab binding in an area would be to show correspondingly lower vascular density or a lack of perfusion in existing vessels; however, analysis of microvessel density and the fraction of perfused vessels displayed no correlating trends through tumor tissue (Fig. 5A, i and ii). Intermittent changes to perfusion could affect the localized distribution of trastuzumab for individual vessels, although these temporal changes in blood flow are generally considered to be relatively infrequent and of short duration (33), and given the long half-life of trastuzumab, it is unlikely that intermittent perfusion changes could account for all of the heterogeneity seen with respect to trastuzumab distribution.

Many studies have illustrated the importance of extracellular matrix density and composition in the ability of macromolecules to distribute in the extravascular compartment of tumor tissue (4, 5, 34), although trastuzumab has been shown to have a relatively large interstitial volume of distribution in some tumor models (6). In this study, we examined the vascular extracellular matrix layer, the basal lamina, where the presence of collagen type IV showed consistent association with CD31 throughout the tumor, irrespective of trastuzumab saturation (Fig. 3C). Vascular maturity is also thought to have a significant impact on the stability of tumor vasculature, where pericytes, among other things, regulate vascular diameter and permeability (35, 36). A trend of fewer microvessels associated with pericytes is found in sections located toward the center of tumors relative to proximal sections (Fig. 3D), which corresponds to higher trastuzumab saturation in the same sections. No exploration into this association was done in this study, and these data are not considered evidence of a causal relationship. It is plausible, however, that more mature vessels, as indicated by presence of pericytes, could have improved efficiency in delivery of macromolecules to tumor tissue, and further investigations based on this preliminary data may be warranted.

There are multiple reports supporting the binding site barrier hypothesis first proposed by Weinstein et al., which suggests that high antibody affinity and high expression levels of antigen will hinder the penetration of an antibody through tumor tissue via its own successful binding to antigen (3739). In the present study, examination of trastuzumab at progressive time points in this highly overexpressing HER2 model showed that, with time, trastuzumab is able to reach some cells located far from vasculature. Those vessels associated with poor extravascular distribution of trastuzumab did not exhibit noticeably more or less HER2 expression in perivascular cells, suggesting that the degree of associated trastuzumab is not the result of variable HER2 expression.

Elevated interstitial fluid pressure in tumors has also been suggested as partially responsible for poor monoclonal antibody penetration (3, 40, 41), particularly in the tumor core, where high interstitial fluid pressure principally prevents macromolecular extravasation and also influences interstitial fluid convection. Tumor vessels are otherwise thought to be sufficiently permeable and leaky that the limiting factor for extravasation will be blood flow (42, 43). As also indicated above, Fig. 4 illustrates that at both early and late time points, trastuzumab was found extravasating from blood vessels in the central regions of tumors, as well as at the periphery in a heterogeneous pattern not isolated to large geographic regions, such as the rim or core.

Factors not considered in this study include rate of blood flow, vascular permeability, and tumor tissue interstitial components that would affect convection and the interstitial diffusion coefficient of trastuzumab. Notwithstanding considerable heterogeneity in trastuzumab distribution in the doses and time points examined, the ability of trastuzumab to penetrate through tissue with time seems relatively efficient in consideration of its high molecular weight and high binding affinity. In most studies evaluating the efficacy of trastuzumab, the therapeutic regimen would involve multiple doses and in combination with its long plasma half life, based on these results, it is plausible that trastuzumab would eventually have the opportunity to access most target cells within a solid tumor. Importantly, however, complete saturation of HER2 binding sites throughout the tumor was not seen in our study, and it is not clear what inhibited its complete distribution. This could have a significant effect on the success of cancer therapy wherein survival of individual cells can lead to tumor repopulation. Further studies examining distribution of trastuzumab after prolonged therapy and investigation of relative changes in pharmacodynamic markers between regions with and without bound trastuzumab could provide clarity regarding whether poor tumor tissue penetration is a mechanism for resistance to trastuzumab.

Grant support: Canadian Breast Cancer Foundation and Canadian Institutes of Health Research. J.H.E. Baker and L.A. Huxham are Michael Smith Foundation for Health Research scholars. A.H. Kyle is a Michael Smith Foundation for Health Research and Canadian Institutes of Health Research scholar.

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.

Note: Jennifer H.E. Baker and Kirstin E. Lindquist contributed equally to this article.

The authors wish to gratefully acknowledge Drs. Corinna Warburton and Wieslawa Dragowska for providing us with the LCC6 tumor models.

1
Minchinton AI, Tannock IF. Drug penetration in solid tumours.
Nat Rev Cancer
2006
;
6
:
583
–92.
2
Carmeliet P, Jain RK. Angiogenesis in cancer and other diseases.
Nature
2000
;
407
:
249
–57.
3
Jain RK. Physiological barriers to delivery of monoclonal antibodies and other macromolecules in tumors.
Cancer Res
1990
;
50
:
814s
–9.
4
Baluk P, Morikawa S, Haskell A, Mancuso M, McDonald DM. Abnormalities of basement membrane on blood vessels and endothelial sprouts in tumors.
Am J Pathol
2003
;
163
:
1801
–15.
5
Netti PA, Berk DA, Swartz MA, Grodzinsky AJ, Jain RK. Role of extracellular matrix assembly in interstitial transport in solid tumors.
Cancer Res
2000
;
60
:
2497
–503.
6
Wiig H, Gyenge CC, Tenstad O. The interstitial distribution of macromolecules in rat tumours is influenced by the negatively charged matrix components.
J Physiol
2005
;
567
:
557
–67.
7
Coussens LM, Werb Z. Inflammation and cancer.
Nature
2002
;
420
:
860
–7.
8
Menard S, Casalini P, Campiglio M, Pupa S, Agresti R, Tagliabue E. HER2 overexpression in various tumor types, focussing on its relationship to the development of invasive breast cancer.
Ann Oncol
2001
;
12
Suppl 1:
S15
–9.
9
Borst P, Evers R, Kool M, Wijnholds J. A family of drug transporters: the multidrug resistance-associated proteins.
J Natl Cancer Inst
2000
;
92
:
1295
–302.
10
Primeau AJ, Rendon A, Hedley D, Lilge L, Tannock IF. The distribution of the anticancer drug Doxorubicin in relation to blood vessels in solid tumors.
Clin Cancer Res
2005
;
11
:
8782
–8.
11
Huxham LA, Kyle AH, Baker JH, Nykilchuk LK, Minchinton AI. Microregional effects of gemcitabine in HCT-116 xenografts.
Cancer Res
2004
;
64
:
6537
–41.
12
Kyle AH, Huxham LA, Yeoman DM, Minchinton AI. Limited tissue penetration of taxanes: a mechanism for resistance in solid tumors.
Clin Cancer Res
2007
;
13
:
2804
–10.
13
Holliger P, Hudson PJ. Engineered antibody fragments and the rise of single domains.
Nat Biotechnol
2005
;
23
:
1126
–36.
14
Reichert JM, Valge-Archer VE. Development trends for monoclonal antibody cancer therapeutics.
Nat Rev Drug Discov
2007
;
6
:
349
–56.
15
Hudis CA. Trastuzumab-mechanism of action and use in clinical practice.
N Engl J Med
2007
;
357
:
39
–51.
16
Baselga J, Perez EA, Pienkowski T, Bell R. Adjuvant trastuzumab: a milestone in the treatment of HER-2-positive early breast cancer.
Oncologist
2006
;
11
:
4
–12.
17
Mass RD, Press MF, Anderson S, et al. Evaluation of clinical outcomes according to HER2 detection by fluorescence in situ hybridization in women with metastatic breast cancer treated with trastuzumab.
Clin Breast Cancer
2005
;
6
:
240
–6.
18
Piccart-Gebhart MJ, Procter M, Leyland-Jones B, et al. Trastuzumab after adjuvant chemotherapy in HER2-positive breast cancer.
N Engl J Med
2005
;
353
:
1659
–72.
19
Glazyrin A, Shen X, Blanc V, Eliason JF. Direct detection of herceptin/trastuzumab binding on breast tissue sections.
J Histochem Cytochem
2007
;
55
:
25
–33.
20
Adams GP, Schier R, McCall AM, et al. High affinity restricts the localization and tumor penetration of single-chain Fv antibody molecules.
Cancer Res
2001
;
61
:
4750
–5.
21
Grossi PM, Ochiai H, Archer GE, et al. Efficacy of intracerebral microinfusion of trastuzumab in an athymic rat model of intracerebral metastatic breast cancer.
Clin Cancer Res
2003
;
9
:
5514
–20.
22
Dennis MS, Fau-Jin H, Jin H, et al. Imaging tumors with an albumin-binding Fab, a novel tumor-targeting agent.
Cancer Res
2007
;
67
:
254
–61.
23
Palm S, Enmon RM, Jr., Matei C, et al. Pharmacokinetics and biodistribution of 86Y-trastuzumab for 90Y dosimetry in an ovarian carcinoma model: correlative MicroPET and MRI.
J Nucl Med
2003
;
44
:
1148
–55.
24
Kobayashi H, Shirakawa K, Kawamoto S, et al. Rapid accumulation and internalization of radiolabeled herceptin in an inflammatory breast cancer xenograft with vasculogenic mimicry predicted by the contrast-enhanced dynamic MRI with the macromolecular contrast agent G6-(1B4M-Gd)256.
Cancer Res
2002
;
62
:
860
–6.
25
Tada H, Higuchi H, Wanatabe TM, Ohuchi N. In vivo real-time tracking of single quantum dots conjugated with monoclonal anti-HER2 antibody in tumors of mice.
Cancer Res
2007
;
67
:
1138
–44.
26
Trotter MJ, Chaplin DJ, Olive PL. Use of a carbocyanine dye as a marker of functional vasculature in murine tumours.
Br J Cancer
1989
;
59
:
706
–9.
27
Warburton C, Dragowska WH, Gelmon K, et al. Treatment of HER-2/neu overexpressing breast cancer xenograft models with trastuzumab (Herceptin) and gefitinib (ZD1839): drug combination effects on tumor growth, HER-2/neu and epidermal growth factor receptor expression, and viable hypoxic cell fraction.
Clin Cancer Res
2004
;
10
:
2512
–24.
28
Rae J, Creighton C, Meck J, Haddad B, Johnson M. MDA-MB-435 cells are derived from M14 Melanoma cells-a loss for breast cancer, but a boon for melanoma research.
Breast Cancer Res Treat
2007
;
104
:
13
–9.
29
Lub-de Hooge MN, Kosterink JG, Perik PJ, et al. Preclinical characterisation of 111In-DTPA-trastuzumab.
Br J Pharmacol
2004
;
143
:
99
–106.
30
Tang Y, Scollard D, Chen P, et al. Imaging of HER2/neu expression in BT-474 human breast cancer xenografts in athymic mice using [99mTc]-HYNIC-trastuzumab (Herceptin) Fab fragments.
Nucl Med Commun
2005
;
26
:
427
–32.
31
Baselga J, Norton L, Albanell J, Kim YM, Mendelsohn J. Recombinant humanized anti-HER2 antibody (Herceptin) enhances the antitumor activity of paclitaxel and doxorubicin against HER2/neu overexpressing human breast cancer xenografts.
Cancer Res
1998
;
58
:
2825
–31.
32
Mandler R, Kobayashi H, Hinson ER, Brechbiel MW, Waldmann TA. Herceptin-geldanamycin immunoconjugates: pharmacokinetics, biodistribution, and enhanced antitumor activity.
Cancer Res
2004
;
64
:
1460
–7.
33
Durand RE. Intermittent blood flow in solid tumours-an under-appreciated source of ‘drug resistance.’
Cancer Metastasis Rev
2001
;
20
:
57
–61.
34
Choi J, Credit K, Henderson K, et al. Intraperitoneal immunotherapy for metastatic ovarian carcinoma: resistance of intratumoral collagen to antibody penetration.
Clin Cancer Res
2006
;
12
:
1906
–12.
35
Jain RK. Molecular regulation of vessel maturation.
Nat Med
2003
;
9
:
685
–93.
36
Morikawa S, Baluk P, Kaidoh T, Haskell A, Jain RK, McDonald DM. Abnormalities in pericytes on blood vessels and endothelial sprouts in tumors.
Am J Pathol
2002
;
160
:
985
–1000.
37
Weinstein JN, van Osdol W. Early Intervention in Cancer Using Monoclonal antibodies and other biological ligands: micropharmacology and the “binding site barrier.”
Cancer Res
1992
;
52
:
2747s
–51.
38
Saga T, Neumann RD, Heya T, et al. Targeting cancer micrometastases with monoclonal antibodies: a binding-site barrier.
Proc Natl Acad Sci U S A
1995
;
92
:
8999
–9003.
39
Juweid M, Neumann R, Paik C, et al. Micropharmacology of monoclonal antibodies in solid tumors: direct experimental evidence for a binding site barrier.
Cancer Res
1992
;
52
:
5144
–53.
40
Flessner MF, Choi J, Credit K, Deverkadra R, Henderson K. Resistance of tumor interstitial pressure to the penetration of intraperitoneally delivered antibodies into metastatic ovarian tumors.
Clin Cancer Res
2005
;
11
:
3117
–25.
41
Jain RK, Baxter LT. Mechanisms of heterogeneous distribution of monoclonal antibodies and other macromolecules in tumors: significance of elevated interstitial pressure.
Cancer Res
1988
;
48
:
7022
–32.
42
McDonald DM, Baluk P. Significance of blood vessel leakiness in cancer.
Cancer Res
2002
;
62
:
5381
–5.
43
Yuan F, Dellian M, Fukumura D, et al. Vascular permeability in a human tumor xenograft: molecular size dependence and cutoff size.
Cancer Res
1995
;
55
:
3752
–6.