compound

Field of invention

The invention relates to the testing of a material response, for instance a tissue response, with respect to compounds such as drugs. It more precisely relates to a device and a method, which can be used for such an assessment.

Background

Determining the sensitivity of a patient to a specific compound and dose prior to its administration is of great value given the high variability among patients observed in clinics today. For example, only 23% of all metastatic colorectal cancer patients that are treated with a combination of cetuximab and irinotecan have shown a partial or complete response, although more than 65% of them suffered from severe (grade III) to life-threatening (grade IV] adverse effects ¹ . This variability in drug sensitivity is due to the unique cellular and molecular properties of every tissue implying that information about drug effects can be gained reliably only when testing the compound in question on the target tissue itself.

Once the tissue is resected from the patient's body during a biopsy or surgery, it has become a common practice to suspend the cells that constitute the tissue. This has the advantage of maintaining and manipulating them with ease as well as increasing their number to multiply the amount of tests that are performed in consequence. However, this ultimately also leads to the destruction of the tissue microenvironment with its specific structure and organization, its loss of stromal-epithelial interaction and cellular heterogeneity. Accordingly, the response of a tissue towards a compound may change considerably ² . For example, intercellular communication between nearby cells plays an important role within the tumor microenvironment where immune and epithelial cells produce molecules (e.g. cytokines) that are recognized by cancer cells and that induce changes within them. However, such paracrine effects are mostly absent when the tumor microenvironment is destroyed and cell heterogeneity is lost due to the inevitable selection process that takes place when sub-culturing the cells.

Accordingly, efforts were put into finding ways to maintain tissue samples intact outside the human body [ex vivo). Despite of the fact that blood flow through the tissue is stopped once removed from the human body, it was shown that the culture of l-3mm ³ fragments is doable when embedded within a collagen gel matrix ^3-5 , placed on top of a gelatin foam ^{6 7} or cultured on a Teflon membrane ⁸ . In all cases, the tissue could be kept alive for several days. Nevertheless, it remains challenging to expose a tissue correctly to a compound whose effect on the tissue is to be studied. A substance that is added to the surrounding solution has to penetrate into the tissue. In fact, long diffusion distances result in long diffusion times and transient concentration gradients between the periphery and the inside of the tissue. Consequently, the region that is treated most efficiently with the drug is also the one that is mostly damaged from the excision process, i.e. the tissue periphery. The tissue inside, on the other hand, will be exposed with an unknown concentration- time profile.

The detection of compound effects on the tissue represents a further challenge. Direct optical microscopy is limited by relatively thick tissue samples. Therefore, histology has become one standard where the tissue is fixed, embedded in paraffin and stained for a variety of molecules using dyes (histochemistry), radiolabels (autoradiography) or antibodies (immunohistochemistry). However, besides being resource consuming, this technique requires the experiment to be stopped for analysis and consequently does not allow acquiring multiple data points in time. Similarly, enzymatic assays, which measure the metabolic activity by detecting molecules within the tissue-surrounding medium, may have detection limits that are exceeded after a minimum time of culture only. Again, this results in a limited amount of information on the same sample. Moreover, most assays have relatively complicated protocols and require trained personnel. This rules out the possibility for a standard and renders these tests prone to human error and misinterpretation. Furthermore, metabolic endpoints depend on the precise number of cells within the tissue, a parameter that may change from one tissue to another and from one region within the tissue to another. These are important drawbacks of current practices since in most cases the volume of tissue obtained during a biopsy is relatively small, and therefore few information can be obtained about the kinetics of compound effects for example. Most importantly, the choice of which information to look for and thus the decision which test to perform further complicates the problem.

The maintenance of a tissue's microenvironment together with its cell type heterogeneity is essential when studying compound effects. However, there are major challenges associated to the thickness of these tissues, i.e. exposing the tissue to the compound correctly and monitoring the induced effect within the tissue, i.e. at the same location where the compound has been delivered.

Existing testing methods are briefly described below.

Drug treatment of ex vivo tissues

Three strategies have been pursued in the past to render the exposure of a tissue to a drug more efficient. On the one hand, the tissue thickness can be decreased by cutting it into very small fragments (1mm ³ or smaller) ^{3 5} or by slicing it using a razor blade (typically 300um thick) ^8-11 . However, the later involves the immobilization of the tissue on a support (using a glue) within an ice-cold solution. Thus the slicing as well as the manipulation of the resulting slice requires additional time and technical know-how and ultimately represents a stress on the tissue that should not be underestimated. Regarding tissue fragments, it remains unclear to what extent cells in the core of the tissue fragment remain viable ¹² and are impacted by a substance added to the surrounding solution.

It was further suggested to improve mass transfer by introducing dynamic conditions, for example by culturing tissue samples within a rotating micro-bioreactor where the tissue sample is maintained suspended in a free fall condition induced by gravity and within a low turbulence regimen (US5437998 A).

A third method to optimize the drug exposure of a tissue is, as was mentioned above, by decomposing the tumor environment into individual cells. Cells can be cultured in vitro in monolayers. Consequently, every cell will experience the same drug concentration and diffusion effects become negligible. However, it was shown previously that the response of cells cultured in vitro towards a compound differs from the one observed in vivo and therefore this method exhibits only poor clinical relevance.

Passive electrical properties of ex vivo tissues

The response of a tissue following its exposure to a compound might be very complex. The measurement of the passive electrical properties of a tissue (bio-impedance) allows integrating this complexity of the cellular signaling network by summing it up into a detectable response parameter. Moreover, impedance measurements are label-free and non-destructive allowing 1) for the continuous acquisition of a large number of time points and thus for the study of compound effect kinetics and 2) for the study of thick tissue samples due to its large penetration depth.

Although many studies on ex vivo cultures have been conducted so far, impedance spectroscopy has rarely been used as an analysis method. Instead, impedance-based screening has been carried out on monolayer cultures where cells are grown on an electrode implemented on the bottom of a culture dish ^{13 14} . As was mentioned above, monolayer and mono-cell type culture poorly correlates with the response observed in tissues ² . Researchers have also cultured tissue samples directly on such planar electrodes ^{10 11} , however, these do not attach well on the electrode surface and therefore do not provide reliable experimental results. This is due to tissue movement and leak currents between the tissue and the substrate, both leading to a lowered sensitivity for trans-tissue impedance effects (extra- and intracellular current paths).

Different approaches using impedance-based sensors have been suggested in literature to study conditions in tissues. Robitzki et al. introduced a setup that comprises the culture of a tissue on top of a liquid permeable membrane, which separates a bottom from a top chamber with each of them containing one electrode (US20090247898). The impedance is measured across the membrane and thus partly across the tissue. This technique is simple, however, it does carry several drawbacks. The measured impedance is strongly dependent on the tissue size and position on the membrane resulting in a lowered reproducibility. Furthermore, leak currents bypassing the tissue will dominate the impedance measurement leading to a low sensitivity for changes occurring within the tissue. By consequence, the tissue size and the outer tissue layer, which is damaged the most from resection, determine the measured response parameter. Moreover, the tissue sample is hardly immobilized and can move which causes the impedance signal to be unstable, especially when a drug is added or the medium is changed.

A different approach was suggested in the past that circumvents the problem of leak currents. Metallic probes have been implanted into living animals for monitoring different tissues after pentobarbital-induced respiratory and cardiac arrest ¹⁵ . However, the drug effect was detected indirectly with the administration of the compound at a distinct location within the tissue or within the living animal and therefore tissue properties are not captured where the drug effect is strongest. Such a strategy is similar to adding a compound to the medium surrounding the tissue in an ex vivo configuration. Comparable implants for impedance measurements have been employed for locating a specific tissue and positioning of a needle within that tissue (US20090036794, US6337994 Bl, US6709380 B2) as well as for detecting cancer tissue (US8540710 B2, US20070191733 Al, US20030138378 Al). Although these devices significantly increase the sensitivity for tissue properties and allow screening the tissue inside, they have been limited to diagnostic/localization purposes with one-time measurements (space evolution).

Overall, the continuous measurement of the temporal evolution of drug effects on tissues at the site of action using tissue-penetrating probes remains a major issue.

There are two inter-dependent challenges associated to the investigation of compound effects on tissue explants, i.e. the correct exposure of the tissue interior to the compound and the monitoring of compound-induced tissue changes precisely where the drug acts. Efforts have been mainly put either on removing tissue matter (slicing, tissue dissociation) in order to make the tissue thinner or on optimizing the liquid-tissue contact (dynamic flow conditions, gels, membranes). However, when electrodes are situated outside the tissue, currents tend to bypass the tissue, which renders the measurement very much sensitive to tissue size, movement and tissue-electrode contact. On the other hand, when the impedance is screened from the tissue inside (e.g. by using a probe), the drug barely has access and therefore the impedance is not measured where the drug effect is strongest or only with a considerable lag time (retardation effect). There is a need for a solution that can correctly expose a tissue of any size and shape, out- or inside a living organism, to a compound (condition creation) and assess, over time, reliably the impedance of the compound effect precisely where the effect is strongest (condition detection).

General description of the invention

The previous cited problems are solved with the device and the method defined in the claims. An essential feature of the invention is to provide the same location for the impedance measurement and the compound action site.

The device and method according to the invention may be efficiently used for testing a tissue response to a compound such as a drug.

The invention is however not limited to this field. It may also be used for testing the response to a compound of any material for which the impedance can be measured. The material can be a living or an inert one.

In a preferred embodiment the impedance measuring means and the compound delivering means are contained in the same body.

Advantageously the body comprises a hollow probe through which the compound may be administrated and which wall contains at least one conductive layer for the transmission of the impedance electrical current.

Detailed description of the invention

The invention will be better understood below, in a general description together with the presentation of some embodiments. Some figures are also provided to illustrate the invention.

Brief description of the figures :

Figure 1: Concept of the perfusion and impedance zone localization

Figure 2: Impedance flow sensor

Figure 3: Impedance-based bulk elasticity measurement

Figure 4: Probe and tissue container for ex vivo monitoring of a tissue fragment

Figure 5: Support for the probe for ex vivo monitoring of a tissue fragment

Figure 6: Card and reader unit

Figure 7: Incubator configurations Figure 8: Flow configuration

Figure 9: Compound exposure configuration

Figure 10: Electrode configuration

Figure 11: Probe shape

In a general manner (see figure 1) the present invention relates to the exposure of a zone (3) within a tissue (1) to a compound and the simultaneous impedance assessment of the compound effect on the tissue precisely within the exposed zone (3). To this scope, a probe (4) may be advantageously used to penetrate into a tissue (1); where the penetration can extent to less than 1% or more than 100% of the entire tissue thickness. The tissue (1) can be inside [in vivo) or outside [ex vivo) the living organism. The probe section can be as small as lOnm and as large as several centimeters. The penetration depth of the probe (4) into the tissue (1) can be as small as one nanometer and as long as several centimeters. The probe is preferably hollow (2) allowing for the delivery and perfusion of the tissue interior with a solution or gas containing or not a compound of a certain concentration. The pressure or flow imposed on the hollow part of the probe (2) can lead to an interstitial flow expanding from the probe opening (8) into the tissue (1). The imposed pressure or flow can also create a liquid path along the probe allowing for a stronger ejection of solution/gas out of the probe opening (8). The probe shape might be of importance to whether the flow occurs mainly across the tissue or along the probe (4). Also, the probe (4) may exert a negative pressure to aspirate the tissue against the probe to increase measurement sensitivity and/or hinder the flow of the liquid along the probe (4) and favor liquid penetration into the tissue (1). The probe (4) remains either inside the tissue (1) for the entire treatment length, which can be as short as less than a second and as long as several weeks, months or years, or the probe (4) is removed between two time points and re-inserted only for the measurement. Advantageously the probe is either completely or partially made of an electrically conducting material (5). The electrically conducting material (5) may be electrically insulated using an insulating material (6) except near or at the fluidic opening of the probe [7, 8). The non-insulated part of the probe (7) is used to assess the impedance of the zone of the tissue (1) that is exposed to the compound. The impedance measurement is localized since the energy dissipation is constricted to a zone close to the non-insulated part of the electrical conductor, i.e. the electrode (8) (mono-polar impedance measurement setup). The surface area of the electrode determines the extent of localization (9).

Accordingly, the compound-exposed cells are the ones which contribute most to the impedance measurement since the electric field concentrates around the electrode (7), which is equivalent to or near the fluidic opening (8) (perfusion and impedance zone localization). Therefore, lag times originating from the diffusion of the compound to the zone of highest impedance sensitivity (9) are avoided. By consequence, the effect of the compound as well as the assessment of the impedance is localized and these two localizations are almost equivalent. The invention further comprises an impedance flow sensor, which exploits the above-mentioned superposition of the zone of injection and zone of impedance measurement to detect flow precisely at the probe opening. A flow test is important to check the ability of a specific probe (4) to eject liquid by comparing whether the obtained flow at the probe opening (8) corresponds to the set flow. In fact, this may be crucial since a probe (4) can be partially or completely clogged (e.g. dust, salt deposits, manufacturing fault), which would falsify the experimental results. Two modes of operation are described hereafter: (1) A solution with a certain electrical conductivity is injected into a bulk solution or tissue with a different electrical conductivity (bulk conductivity unequal ejecta conductivity). At start, the probe in- and outside have the same conductivity as the bulk solution (figure 2a, phase 1). With flow, the conductivity of the close proximity (dotted line in figure 2a) and the inside of the probe will change (figure 2a, phase 2), thus resulting in changed impedance (figure 2b and c). The conductivity of the bulk volume on the other hand does not change considerably since its volume is much larger than the one of the ejected liquid. When the flow is stopped, the bulk solution diffuses to the close proximity of and inside the probe, thus changing the conductivity to the one of the bulk. This reflects in an impedance change in the reverse direction (figure 2b and c). Hence, an injection in intervals represents a good mean to check on the ejection capacity of the probe (4), however, it can also be used in the continuous flow mode (e.g. serial perfusion of solutions with different conductivity). The resolution of this sensor is determined, amongst others, by the cross section of the probe opening and is therefore mostly limited by the precision of the probe fabrication. This mode of function can also work in the reverse direction by aspiration. It is further stated that the frequency at which the impedance is measured shifts the measurement more to the probe in- or outside. In fact, the probe inside contributes more to the measured impedance at lower frequencies and the probe outside at higher frequencies (equivalent electrical circuit in figure 2d). (2) The tissue is pushed away from the probe opening once the probe is inserted into the tissue and liquid or gas is ejected (figure 3). This results in decreased impedance. Given its elasticity, the tissue moves back against the probe when the flow is stopped, leading to an impedance increase. This phenomenon can be exploited to check the ability of the probe to eject liquid during the experiment when the conductivity of the ejecta and the bulk are equal.

The above-described probe can further be used to characterize the mechanical properties of an elastic bulk (e.g. tissue). When a solution or gas is ejected from the probe (4), the tissue (1) is pushed away (figure 3b), reflecting in lowered impedance. Once the flow is stopped, the elastic material or tissue (1) comes back (figure 3a) and thereby increases the impedance. The kinetics of the back and forth movement is characterized by the elastic properties of the material or tissue. This principle also works in an aspiration mode.

This invention further involves the combination of the described impedance measurement with the measurement of different metabolic parameters (e.g. pH, glucose, lactic acid, CO2). For example, pH measurements allow for the quantification of aerobic and anaerobic respiration of the cells constituting the tumor tissue. The resulting acidification strongly depends on the number of cells present within the tissue and therefore should be normalized to the cell density, which can be quantified by measuring the impedance with the probe (4). Furthermore, the impedance measurement that is performed by the probe (4) can provide information on whether the tissue is cancerous or normal and whether it is alive or dead at the beginning of the test. These results will relativize the data obtained from metabolic endpoints and thereby put them into perspective. Consequently, all parameters (impedance and metabolic) can be integrated to obtain a more reliable test result.

Preferred embodiment

In a preferred embodiment, the invention provides a system to detect drug-induced changes on a patient's own tissue outside the body [ex vivo) in order to predict the tissue response and accordingly to define the drug therapy. To this scope, the probe (4) is immobilized within a support (10), from which it sticks out by a distance d2 (figure 4). This precise distance depends on the thickness of the overlayed tissue (di + d2) such that the probe always penetrates up to a specific depth into the tissue, preferably by half the tissue thickness (d2). Accordingly the probe opening (8) and electrode (7) are entirely located within the tissue.

The probe opening (8) is located on the side of the probe in order to prevent the clogging of the opening when inserted into the tissue (1). The tip end of the probe (11) is closed and pointed. Pencil point probes are suitable for piercing without coring. Accordingly, the tissue will not be cut, as it would be with a sharp bevel needle. By consequence, the tissue-probe contact will be stronger, thus impeding liquid leakage along the probe (4).

When using tissue samples as small as 1mm ³ , it becomes difficult to center the tissue on the probe and to prevent the tissue from floating within the surrounding medium. Therefore, this embodiment involves a container (12) within which the tissue is put prior to placing it onto the probe (4). More precisely, the tissue (1) is placed into the hollow part of the container (13), which is of similar size as the tissue (1). The container with the tissue is then placed into a cavity (14) within the support (10) to x-y align the tissue with regard to the probe and to prevent the tissue from moving in x-y direction after placement. Furthermore, the container (12) prevents the tissue from floating once tissue culture medium is added to the larger well. This is achieved when the container (12) is forced towards the bottom of the cavity either through gravity (container material has a higher density than the medium) or through other forces (e.g. magnetic force, where the container and the support are provided with small magnets; weight from the top; suction through negative pressure from the bottom).

The tissue container (12) has holes at the top (15) to allow the tissue (1) to be in contact with the surrounding medium. Eventually, the container additionally has bottom escape channels (16) through which liquid evacuates when the container (12) is placed into the cavity (14) and when the tissue (1) is perfused.

The probe support (10) further comprises a larger well (17) (figure 5) that is used to immerse the tissue (1) and the container (12) in liquid. The well (17) may consist of a conductive material such that it can be used as a large counter electrode. The support (10) may also contain a channel (18) to provide liquid to openings (19) within the probe (4) in order to access the hollow part (2) of the probe (4) and the opening (8) of the probe (4). The probe (4) is further equipped with an electrical connector (20) for adapting the small probe (4) for use with a larger port.

Multiple devices (21), each comprising a support (10) with a probe (4), are assembled to an array on a card (22). This card is connected to a reader unit (23), which contains all electronic and fluidic parts necessary to perform the test (figure 6).

All devices are maintained at a certain temperature, gas concentration (CO2, O2) and humidity. In a preferred embodiment, this is achieved by using an incubator (27) that is part of the reader unit and/or the card (figure 7b). This one is heated by using a warm air blower, electrical resistances or circulating pre-heated liquid. Alternatively, the whole setup is either placed into a larger incubator (figure 7a) or every single device is equipped with its own small incubator (figure 7c). The drug delivery to the solid tissue (1) may be performed at low flow rates, generated by using a pressure (Δρ) or flow (AV) pump (24) (figure 6, figure 8). A flow sensor (28) may be employed to detect the flow rate and feedback-control the pump (24). The perfusion can have different profiles (e.g. continuous or in intervals/i) in order control the amount of liquid ejected into the tissue and to account for the limited capacity of the tissue to uptake liquid.

The impedance is recorded using an impedance analyzer (25). This analyzer records the impedance of the tissue (1) continuously allowing for the analysis of the kinetics of the compound effect. Furthermore, the impedance is measured at different frequencies, allowing for the detection of extra- and intracellular phenomena. Those have been related to minor cell damage (morphology changes) and major cell injury (cell death) before ^{16 17} .

Based on the kinetics of the tissue changes induced by the compound, at least four parameters can be made available to the end user (e.g. physician) through the user interface (26), i.e. Tox(AR)minor, Tox(AR)major, Tminor, Tmajor. Those parameters may be sum up to a ranking score parameter, which indicates to the physician the success chance of a therapy for a specific patient when using the tested compound(s). Although the machine will provide a synthesized version of the original acquired data (Tox(AR), τ or Ranking Parameter TOX), it will be possible for the physician to access all the data at any time if desired.

In the preferred embodiment, the tissue explant is cut and split into individual samples, each of them being exposed to a condition (figure 9a and b). A condition can be a compound of a certain concentration (figure 9a), a mixture of compounds (figure 9b), specific culture conditions (e.g. gas pressures, perfusion, medium composition) or a control with culture medium only.

Further embodiments

In a second embodiment, several probes are immobilized in one support and penetrate into the same tissue sample simultaneously (figure 9c and d). All probes can perfuse the same (figure 9c) or a different compound/mixture of compounds (figure 9d). In a further embodiment, the tissue is not split and a multitude of probes perform the test on the entire tissue explant directly.

In the preferred embodiment, the counter electrode is large and far from the probe electrode (figure 10a). However, the second electrode can be at different positions as exemplified in figure 10. The electrode can be on a second probe, which is immobilized in the support, and that penetrates into the tissue (figure 10b). Both electrodes can be on the same probe (figure 10c). The electrode can also be embedded into a lid (figure lOd) or at the bottom of the well (figure lOe). In a further embodiment, the culture conditions such as the temperature may temporarily be changed either locally using the probe or globally using the incubator in order to study combined hyperthermia-chemotherapy treatment.

In a further embodiment, the probe can have different shapes (figure 11). The probe end can be a blunt end (figure 11a) with different edges (e.g. chamfered, rounded, conical), a sharp end (figure lb) (e.g. conical, beveled, lancet bevel) or a closed end (pencil point, bevel, trocar point). In case of the closed end, the opening is situated on the side (figure 11c). The penetration depth can also exceed the tissue thickness (figure lid). Furthermore, the thickness and shape of the probe can vary along its length in order to strengthen the contact between the tissue and the probe. To this scope, the probe may possess a flat (figure lie), conical (figure llf) or funnel end (figure llg). Also, the entire probe may take a conical shape (figure llh). In all cases, there can be one or multiple openings. The probe may have a porous part through which the liquid is dispensed over a larger surface (figure Hi). The probe may have several openings through which liquid is dispensed, sampled or pressure is applied (figure llj). Furthermore, the hollow part of the probe may be U-shaped with a distal opening (figure Ilk). Accordingly, the compound can be transported very close to the opening without applying any pressure on the tissue. From there it diffuses out of the probe and into the tissue.

In a further embodiment, the presented method may be combined with histology where the tissue is fixed with a fixing agent. The fixing agent is either added to the surrounding medium or perfused inside the tissue through the probe.

The presented method may be applied to a tissue within the living organism. To this scope, a probe is implanted into the patient's target tissue and left for a period of time stretching from a millisecond to many years. The probe can have a form as described in figure 9, but may also be in form of a capsule to limit damage caused by movement. The implantable probe may be equipped such that the signal is sent wirelessly to outside the patient's to avoid wiring.

A further embodiment of the presented invention is its implementation into a biopsy needle or biopsy puncher where the tissue is analyzed for its compound sensitivity directly after its removal from the body.

The invention is of course not limited to the embodiments presented in this document.

The device according to the invention may advantageously be used for testing a tissue response to a compound but, as mentioned previously, it may also be used for testing the response to a compound of any other material for which the impedance may be measured. Reference Numerals

1 Tissue 16 Bottom opening of the container

2 Hollow part of the probe 17 Well

3 Interstitial perfusion/diffusion zone 18 Fluidic channel

4 Probe 19 Probe bottom openings

5 Electrically conductive part of the probe 20 Electrical adaptor

6 Insulated part of the probe 21 Device

7 Electrode 22 Card

8 Probe opening 23 Reader unit

9 Impedance sensitivity zone 24 Flow control

10 Support 25 Impedance analyzer

11 Probe tip 26 Data output

12 Tissue container 27 Incubator

13 Cavity of tissue container 28 Flow sensor

14 Cavity of the support 29 Secondary openings

15 Top openings of the tissue container

References

1. Cunningham, D. & Humblet, Y. Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N. Engl J. Med. 351, 337-345 (2004).

2. Hoffman, R. M. The three-dimensional question: Can clinically relevant tumor drug

resistance be measured in vitro? Cancer Metastasis Rev. 13, 169-173 (1994).

3. Miller, B., Miller, F. & Heppner, G. Assessing tumor drug sensitivity by a new in vitro assay which preserves tumor heterogeneity and subpopulation interactions./ Cell. Physiol. 121, 105-116 (1984).

4. Pirnia, F. et al. Ex vivo assessment of chemotherapy-induced apoptosis and associated molecular changes in patient tumor samples. Anticancer Res. 26, 1765-72 (2006).

5. Freeman, a E. & Hoffman, R. M. In vivo-like growth of human tumors in vitro. Proc. Natl.

Acad. Sci. U. S. A. 83, 2694-8 (1986).

6. Papini, S. et al. Selective growth of epithelial basal cells from human prostate in a three- dimensional organ culture. Prostate 59, 383-92 (2004).

7. Bolenz, C. et al. Topical chemotherapy in human urothelial carcinoma explants: a novel translational tool for preclinical evaluation of experimental intravesical therapies. Eur. Urol. 56, 504-11 (2009).

8. Vaira, V. et al. Preclinical model of organotypic culture for pharmacodynamic profiling of human tumors. Proc. Natl. Acad. Sci. U. S. A 107, 8352-6 (2010).

9. Kern, M. a et al. Ex vivo analysis of antineoplastic agents in precision-cut tissue slices of human origin: effects of cyclooxygenase-2 inhibition in hepatocellular carcinoma. Liver Int. 26, 604-12 (2006).

10. Kleinhans, R. et al. Sensor-based cell and tissue screening for personalized cancer

chemotherapy. Med. Biol. Eng. Comput. 50, 117-26 (2012).

11. Mestres, P., Morguet, A., Schmidt, W., Kob, A. & Thedinga, E. A new method to assess drug sensitivity on breast tumor acute slices preparation. Ann. N. Y. Acad. Sci. 1091, 460-9 (2006). Ferrarini, M. et al. Ex-vivo dynamic 3-D culture of human tissues in the RCCS™ bioreactor allows the study of Multiple Myeloma biology and response to therapy. PLoS One 8, e71613 (2013). Giaever, I. & Keese, C. R. Monitoring fibroblast behavior in tissue culture with an applied electric field. Proc. Natl. Acad. Set U. S. A. 81, 3761-4 (1984). Wegener, J., Keese, C. R. & Giaever, I. Electric cell-substrate impedance sensing (ECIS) as a noninvasive means to monitor the kinetics of cell spreading to artificial surfaces. Exp. Cell Res. 259, 158-66 (2000). Heroux, P. & Bourdages, M. Monitoring living tissues by electrical impedance spectroscopy. Ann. Biomed. Eng. 22, 328-37 (1994). Meissner, R., Eker, B., Kasi, H., Bertsch, A. & Renaud, P. Distinguishing drug-induced minor morphological changes from major cellular damage via label-free impedimetric toxicity screening. Lab Chip 11, 2352-2361 (2011). Eker, B., Meissner, R., Bertsch, A., Mehta, K. & Renaud, P. Label-free recognition of drug resistance via impedimetric screening of breast cancer cells. PLoS One 8, e57423 (2013).