Patent Application
20110052657
This invention relates to implants for administering substances. One embodiment of the invention is especially, but not exclusively, applicable to implants for administering micronutrient, or trace elements or minerals.
Drugs are most often currently administered orally by the ingestion of tablets, capsules or aerosols, or via subcutaneous, intramuscular or intravenous injections or implants. Oral solid dosage forms account for 40-50% of the market, parenteral products—33% and the other more “novel” dosage forms (NDF's), only a few %. There is nonetheless enormous perceived potential for NDF's that can enhance a drug's therapeutic ratio and avoid patient non-compliance. Non-compliance remains a major issue despite 95% of patients being aware of its consequences. Common examples are incomplete courses of antibiotic therapy, using antidepressive drugs for too short a period, and forgetting to take contraceptive pills.
There are known implants that are implanted subcutaneously and which deliver a drug over a period of time in a controlled manner. These are typically based on polymer material systems. There are two basic types of implant for controlled drug delivery; “reservoir” and “monolithic” structures. “Reservoir” devices have layers which are corroded or absorbed by the body to release a depot of drug beneath these control layers. By having successive alternate control layers and drug layers the drug can be released over a period of time. “Monolithic” devices have the drug distributed throughout, so that release kinetics are controlled by slow corrosion and diffusion processes.
Problems include the so-called “burst effect” wherein an unwanted high fraction of the drug is released from the polymer capsule's internal surface quite soon upon in-vivo exposure. Another problem is the continuing need for high-purity, cost effective hosts that are capable of sustained drug delivery over months or years (for some applications).
Other known implants include inert ceramic implants which have the drug held in their pores, the drug having to leave the ceramic implant via a tortuous path of micropores, which delays its release and allows it to be controlled.
This invention concerns slow-release tissue-compatible implants, particularly suited to delivering low payloads of a therapeutic substance to a specific site and/or over a long period of time (“long” may be months or years). Although delivered to the site of the implant the beneficial substance may be taken up by the body globally, and may have an effect at another site. In the past the major limiting factor for most drug delivery systems that use implanted materials (polymers or ceramics) has been the “payload” achievable. With the advent of new genetically engineered, more potent drugs, (peptides, proteins, DNA fragments), miniaturised delivery systems are becoming more and more attractive, provided designs ensure patient safety. An example of such a safety issue for in-vivo administration would be the failure of an electronic “gate” linked to a large on-chip liquid reservoir. Such concerns can be addressed by the use of drug delivery arrays or drug incorporation within a resorbable host material.
The invention also concerns impregnation of porous semiconducting materials including porous silicon. It is advantageous to have implants comprising porous semiconducting materials that have been impregnated with one or more beneficial substances. It is also advantageous to have the concentration of such a substance or substances as high as possible and as deep as possible from the surface of the porous semiconducting material. A problem with prior art methods of impregnation, for example those disclosed in the paper entitled: “Impregnation of porous silicon” by R Herino (EMIS datareview on porous silicon (1997)) p 66 or the paper entitled “Quenching of porous silicon photoluminescence by deposition of metal adsorbates” by D Andsager, J Hilliard, J M Hetrick, L H Abu Hassan, M Pilsch and M H Nayfeh, reported in J. Appl. Phys. (1993), 74, 4783, is that the depth of impregnation is very low, typically a few atomic percent at 300 nm or less.
Embodiments of the invention will now be described by way of example only, with reference to the accompanying drawings, in which:
According to a first aspect the invention comprises a silicon implant provided with a substance to be administered to the implanted subject.
Preferably the implant comprises porous silicon. The porous silicon may have a porosity of at least 2%, 3%, 4%, 5%, 10%, 20%, 30%, 40%, 50%, 60%, 70%, 80%, or higher (the porosity is the fractional void content by volume). The porous silicon may have a porosity that is in a range between any two of the figures mentioned above. For example, the porous silicon may have a porosity between about 2% and 80%, or between about 4% and 50%, or about 30%.
The implant may have a coating, region, or layer of silicon, or it may be silicon substantially throughout its cross-section. The implant may have a layer of material over the silicon, for example a coating of hydroxyapatite. The over-layer of material may have a physiological effect upon implanting of the implant.
The silicon may be polycrystalline silicon.
Said substance may be distributed in the solid phase silicon material substantially uniformly. In the case of porous silicon said substance may be distributed in the pore network and/or in the silicon skeleton. It is envisaged that distributing the substance in the material of the skeleton may give greater control over the release rate of the substance since this will be related directly to the erosion rate of the silicon material. With a substance held in pores the release rate is also dependent upon how quickly the material can escape the pores (before the skeleton has eroded). This may or may not be desirable or acceptable. In the case of polycrystalline silicon the substance could be distributed in the grains and/or grain boundaries.
It has been appreciated that silicon, and in particular porous silicon, has very good properties which enable it to serve as a drug or micronutrient delivery vehicle. Experimental evidence in support of the suitability of porous silicon as a substance delivery vehicle in an implant has been obtained. Studies by the inventors have shown that porous silicon is “resorbable” or “bio-erodible”, and is resorbed or eroded by the mammalian body at a slow enough rate to make long-term porous silicon implants a practical way to deliver drugs/substances.
Highly porous silicon has long been known to be unstable structurally and chemically, and researchers in the opto-electronics field have gone to great lengths to make it more stable for opto-electronic applications. It is ironic that the lack of stability/inert properties of porous silicon now, with hindsight, is a factor in the controlled delivery of substances by implants.
Tests show that high porosity (e.g. 80%) silicon is resorbed faster than medium porosity (e.g. 50%) silicon, which is in turn resorbed faster than bulk silicon (which shows little, if any, signs of being resorbed). Thus, by adjusting the pore size and total volume of pore to skeleton in porous silicon it is possible to tune the silicon material to be resorbed faster or slower.
Microporous silicon (pore size less than 2 nm), mesoporous silicon (pore size 2-50 nm) and macroporous silicon (pore size >50 nm) are all suitable carrier materials for erosion.
Silicon is cheap, and is available in very pure forms (e.g. the electronics industry already has a requirement for clean, pure, silicon wafers). Moreover, it is already known how to dope silicon crystals with a very wide range of elements, albeit in a different field and albeit at very low concentrations (lower concentrations than required for micronutrients).
It is envisaged that having a beneficial substance provided in porous silicon implant as a delivery mechanism will be especially appropriate for delivering substances which do not need to be delivered at high doses. It is envisaged that a porous silicon implant may be about 0.5×0.5×4 mm in size (or in the ranges >0 to 2 mm×>0 to 20 mm×>0 to 20 mm). Each implant may have a weight of less than a milligramme, or a few milligrammes, or a few tens or hundreds of milligrammes, and each tablet may conveniently be doped with a “dry payload” of tens to hundreds of microgrammes of substance, or even to a few milligrammes (or more if it is possible to carry it). This may be insufficient for delivery of macronutrients, or macro dose drugs, but it is sufficient to deliver substances which are needed in the micro to milligramme range.
One area where porous silicon is suitable as a carrier for a therapeutic or beneficial substance is in the provision of micronutrients or microminerals to subjects.
Some trace minerals needed by the body are present at extremely low concentration in the body (e.g. selenium, chromium, manganese and molybdenum). The recommended daily allowance (RDA) of microminerals can be <0.1 mg/day and yet deficiency effects are well documented (e.g. selenium iodine). This is often because only a small and highly variable fraction of orally-ingested microminerals are absorbed and hence become bio-available. Having these microminerals delivered by an implanted silicon tablet that is fully adsorbed is an attractive solution to deficiency problems. Moreover, by having the substances in an implant it is possible to deliver a substance to a specific site (e.g. iodine to or near the thyroid gland).
Silicon itself is an essential trace element, and a porous silicon implant could, of course, be used to deliver silicon in which case it may not carry any other beneficial substance.
The implant may have more than one beneficial substance. A multi-essential trace element implant having 2, 3, 4, 5 or more trace elements may be provided.
Other elements have widespread use clinically for therapeutic purposes, e.g. lithium for depression, gold and silver for antibacterial properties, and platinum for neoplastic diseases. These may be administered not so much to achieve a desirable “normal” mineral content in the physiology of a subject, but to increase levels of microminerals to therapeutic levels, possibly at a specific locality. The dose levels of such therapeutic elements in the blood stream are normally in the μg/l range, which is within the capabilities of porous silicon implants. The implant may comprise a porous silicon sample in which such an element (whether it be a trace element or an element in a beneficial substance or some other element of the periodic table) has been impregnated at a concentration between 1 and 90 atomic percent at a depth, from the surface of the sample, between 0.35 μm and 1000 μm. More preferably the element may be present at a concentration between 1 and 90 atomic percent at a depth, from the sample surface, between 1 μm and 1000 μm. Yet more preferably the element may be present at a concentration between 1 and 90 atomic percent at a depth, measured from the sample surface, between 10 μm and 1000 μm. Even more preferably the element may be present at a concentration between 30 μm and 1000 μm. It is often advantageous for such elements to be released into the body of the mammal at a slow rate. To facilitate this slow release it is advantageous to provide high concentrations of such elements at relatively large depths from the surface of the porous silicon.
A therapeutic or essential trace element (or other element) may be delivered in non-elemental form. For example, a salt of a metal may be the beneficial substance, metal ions being available to the body of the patient. So long as the substance is delivered in a physiologically usable form, how it is carried in the erodable material (compounded or elemental) may not matter.
It will be appreciated that implanting an implant which can deliver a controlled amount of a drug/micronutrient/micromineral for a month, or even two or three months, or a year, or possibly even years, has great advantages over relying on a patient to eat correctly or take oral tablets correctly, especially when the disorder being treated exacerbates any difficulties in the patient having a discipline to take the remedy. The fact that a silicon implant can be made to break down slowly makes it possible to leave an implant alone for a long time. When a sustained level of drug dose delivery is required a silicon implant can be engineered to deliver a prolonged sustained substantially constant (or constant enough for the intended purpose) level of drug or mineral. Using implants to deliver trace elements is attractive for those people who have gastrointestinal tract disorders and who cannot absorb some elements orally. Even if a patient were to be treatable orally there can be a great variation in the level of absorption achieved in people's guts and the same level of oral dietary supplement may result in different levels of absorbed mineral. Subcutaneous absorption has significantly less variation between people and is therefore more easily controlled.
A feature of virtually all drugs, especially large organic molecules, however is that they cannot survive high temperatures. If a silicon implant is made using high temperature doping techniques it may not be possible to get the structural silicon material of the implant to take up some molecules in a functional state. However, this is not a problem for therapeutic elements (e.g. Li, Se, etc.).
Of course, it is possible to use techniques other than thermal drive in to get drugs into the depths of an implant and/or into the solid phase of the porous skeleton. We could use vacuum evaporation, or build the implant up in layers, with the substance adhering predominantly to the surface of each layer, or indeed any suitable technique for distributing the substance throughout the body of the erodable implant.
The geometric design of known monolithic drug release implants can be used to control the drug release rate, and similar techniques of geometric design can be used to control substance release from silicon implants. This may be in addition to controlling the porosity, and pore size of porous silicon to control the rate of dissolution of the implant. The implant may have different porosity at different depths.
Of course, the silicon implant need not necessarily have its erodable carrier material of pure silicon, or substantially pure silicon. Now that it has been established that silicon works it is predictable that silicon carbide and silicon nitride may also have similar properties. Indeed, as a generalisation, a silicon-based compound which has the corrosion properties desired (corroding at a generally constant rate over months or years) and which is non-toxic at the levels released, and which otherwise has no unacceptable harmful effects, would be suitable in place of pure (or substantially pure) silicon as the carrier material, but silicon is still currently preferred.
According to a second aspect the invention comprises a porous or polycrystalline implant provided with a substance to be administered to the implanted subject, the implant being made substantially of an element.
Preferably the implant is made of porous or polycrystalline semiconductor.
Although tests by the inventors in vivo show that porous silicon is corroded if subcutaneously implanted, the inventors also have in vitro tests using simulated body fluid (SBF) which shows that porous and polycrystalline silicon behave in a similar way in SBF.
The realisation that silicon, and especially porous silicon and/or polycrystalline silicon, are suitable materials to be bio-eroded by the body in a controlled manner, and the realisation that this can be used to release drugs/substances into the body (or to a localised area), can be further expanded. The implant may have a plurality of drug payload areas defined in a body of silicon, the body of silicon having a plurality of barrier regions, or doors, adapted to be resorbed in use by the body, the geometry and size of the barrier regions being such that in use at least a first barrier region is eroded or resorbed such that the drug payload in the drug payload area adjacent said first barrier region is released to the body before a second barrier region, adjacent a second drug payload area, is resorbed sufficiently to enable the second drug payload to be released, thereby providing a time-differential breakdown of at least the first and second barrier areas and hence a sequential release of said first and second drug payloads.
The first and second drug payloads may comprise the same drug, or different drugs.
There may be three or more barrier regions each adapted to be corroded at different times, and adapted to release drug payloads from respective drug payload areas at different times.
The barrier regions may comprise porous silicon, such as mesoporous silicon or microporous silicon. The barrier regions may comprise polycrystalline silicon. The rate of erosion of the silicon can be controlled by controlling the porosity (higher porosity materials are corroded faster) and the pore size (smaller pores for same porosity are corroded faster), and the barrier thickness.
Instead of having one implant with a plurality of drug payloads it may be desirable to provide a plurality of separate implants with drug payloads and barrier regions adapted to release the drug payloads at different times.
The implant may comprise a tablet. The tablet may comprise an array of drug payload reservoirs each containing a respective drug payload. The tablet may have a longitudinal direction and the respective barrier regions associated with respective drug payloads may be spaced apart in the longitudinal direction. The implant may be adapted to be corroded in a direction transverse to the longitudinal direction, and preferably at right angles to it, in order to release the respective drug payloads. The implant may have a longitudinally extending surface portion and the drug payload areas may be separated from the surface portion by barrier regions requiring different times of attack by body fluids to be corroded. The different corrosion times of the different barrier regions could be afforded by different thicknesses of barrier region. Alternatively, or additionally, the silicon material of the implant may have different corrosion properties at different barrier regions (e.g. they could be made of porous silicon of different porosity).
According to a third aspect of the invention a method of impregnating a porous semiconducting material with a impregnate substance is provided, the method comprising the step of bringing the impregnate substance into contact with the porous semiconducting material; characterised in that the method further comprises the steps:
It is advantageous for some applications, for example medical applications, to have a method of impregnating substances at a depth of at least several hundred nanometres below the surface of a porous semiconducting sample. It has been found that high depths of impregnation can be achieved by ensuring that the substance to be impregnated is in a molten phase.
Preferably the method of impregnating a porous semiconducting material further comprises the step of thermally decomposing the impregnate substance that has passed into the porous semiconducting material.
Advantageously the method of impregnating a porous semiconducting material comprises the step of reacting the impregnate substance that has passed into the semiconductor material with an oxidant, such as oxygen.
The impregnate substance may be fixed to the porous semiconductor by thermally decomposing it once the substance has entered the interior of the semiconductor. Alternatively the impregnate substance can be fixed to the porous semiconductor by reacting with an oxidant such as oxygen.
Throughout the specification the term “sample surface” is to be taken to mean the surface that separates the sample of porous semiconductor (including porous silicon) from its surroundings. The term does not mean the surfaces that define the pores, unless such surfaces form part of the surface that separates the porous semiconductor from its surroundings.
According to a fourth aspect the invention provides a sample of porous silicon that has been impregnated with an impregnate substance, the sample having a sample surface and the impregnate substance comprising an impregnate element, characterised in that the impregnate element is present at a concentration between 1 and 90 atomic percent at a depth, from the sample surface, between 0.35 μm and 1000 μm.
With reference to the drawings,
The discs used in the in-vivo trials were made as follows:
Titanium foil of 99.6+% purity was purchased from Goodfellow Metals Limited in the form of punched-out discs of 0.5 mm thickness and diameter 11.5 mm. These were subsequently abraded (on both faces) with 12 μm diamond powder to remove any burrs introduced by the punch-out process and to develop equal degrees of surface roughness on both faces. Batches of 10 discs were then chemically etched at a time after cleaning in an ultrasonically agitated acetone bath for 20 minutes. The discs were isotropically etched (to remove surface damage) for 2 minutes in a stirred solution of 35 ml H2O, 10 ml HNO3 and 5 ml 40% HF. The etch process was quenched with DI water and discs were thoroughly rinsed in DI water prior to drying on filter paper.
Batches of 12 mm wide discs were sawn out of ˜5″ (100 mm) diameter CZ wafers (N+, phosphorous-doped 0.0104.0.0156 Ωcm resistivity, 0.5 mm thick, (100) orientation) using a custom-built drill bit. Discs were cleaned in “meths”, then ethyl acetate and then in an ultrasonically agitated acetone bath. Smoothing of the disc edges, removal of saw damage and equality of surface roughness on both faces was achieved using a “polish etch”: 25 ml HNO3+5 ml HF (40%)+5 ml acetic acid. Batches of 10 discs at a time were given a 5-minute etch with continuous stirring followed by a DI H2O quench and rinse with drying on filter paper.
The chemically polished bulk Si discs were anodised sequentially, one at a time, in a custom-built electrolytic cell that enabled both faces and edges to be porosified. Discs were held at one edge by a platinum “croc-clip” and lowered and raised in a controlled manner by a stepping motor into electrolyte in a cylindrical Pt crucible that formed the cathode. Each disc was anodised potentiostatically (i.e. at a constant voltage of 1.0 V) in 40% aqueous HF for a period of 10 minutes. With this type of arrangement most current flow occurs via the meniscus, so the procedure adopted was to slowly raise the meniscus up to the centre of the disc, remove the half-anodised disc, dry it and then invert it, to anodise its other half in the same manner. Fully anodised discs were by this process completely covered in an ˜30 μm thick coating of porous silicon. They were rinsed in DI H2O and dried on filter paper.
All discs were stored in, air prior to sterilisation by “autoclaving” at 134° C. for 10 minutes in pressurised steam.
Tests were also done to ensure that the silicon implants did not cause any serious problems to the guinea pigs, and again these tests show that using silicon, and especially porous silicon, as a biologically acceptable material is viable in a subcutaneous site. The pathological test results are given in the sections, presented below, entitled “Inivo trial of Porous Silicon Implant” and “Score grades” which forms part of this patent application.
The 12-week tests described above have been followed by a 26-week study which has shown entirely consistent results: there is a steady corrosion of porous silicon, and the corrosion of the implants did not cause any significant harmful effects on the test subjects. There was no gross inflammatory response, no significant fibrotic scarring, and excreting the corroded silicon was not a problem.
Since the corrosion of polymer is known and tested as a delivery mechanism for drugs, the present invention is, with the support of the tests discussed, fully realisable. Nevertheless the concept of using semiconductor tablets for prolonged in-vivo drug delivery is completely novel.
The wafer is treated so as to cause it to become porous throughout its depth, and then broken into discrete tablets. The tablets are then smoothed to facilitate subcutaneous insertion and acceptability. An elongate tablet, such as that shown, may be suitable for injection through a needle. The rounded ends 44 of the elongate tablet may help this. The tablets may take the form of that shown in
In an alternative arrangement shown in
It will be appreciated that the implants 42 and 46 are completely eroded in the body and do not need surgically removing.
It is also, of course, envisaged to administer therapeutic elements or substances for particular disorders.
Furthermore, in the case of a problem associated with an excess of an element, or excess of a substance, the implant could be used to administer a blocker to prevent the excess substance operating properly, or something to bind to or react with the excess substance, reducing the effective excess. For example, it has been proposed that silicon in the form of silicic acid can beneficially assist in aluminium excretion.
There is no theoretical reason why an element such as iron cannot be administered using the present invention, but it may as a practical matter be difficult to get a sufficiently high dose of Fe into a silicon tablet to make it sensible to implant an implant.
Elements which are preferred for incorporation into a silicon micromineral tablet include: Vn, Cr, Mn, Se, Mo (dietary requirements), Li, Ag, Au, Pt (therapeutic effect).
As well as silicon other suitable biocorrodable semiconductors to use as carriers for beneficial substances include germanium, silicon carbide and silicon nitride. The semiconductor material could be doped or undoped. Silicon carbide may have anti-thrombogeric properties, and silicon nitride may have orthopaedic applications.
There is no reason why molecules, as well as elements, cannot be delivered, so long as the technique for getting the drug/desirable substance into the implant does not destroy the efficiency of the substance, and so long as they are released in a form which is active when the silicon is broken down.
In the case of minerals/trace elements one way of producing micromineral tablets is:
The order of (1) and (2) may be reversed.
Another way of getting a substance into the implant is to put a salt of a mineral on the surface, heat in an inert atmosphere (e.g. argon) until the material melts and wets the porous structure. The insert/wafer can then be cooled down, and any excess substance washed off in water. A thermal drive in operation can then be performed.
It is preferred to drive the mineral into the solid phase of the porous structure, rather than leave it solely in the pores. This gives greater control of the dissolution rate of the mineral and eliminates the “burst-effect” problem common with polymer-based systems.
A combination of a knowledge of the dissolution rate of the tablets, and how that behaves with time, and the doping level of the tablets, and how uniform that it, gives the ability to control the dosage of substance administered over time.
In the example shown in
By having barrier walls of different thicknesses it is possible to achieve controlled—and sustained—drug release as reservoirs are sequentially opened.
The distances D1, D2 and D3 are such that the “lid” thickness D4 is significantly thinner than the rim thickness D2. Thus, the micromachining of the depths of the reservoirs controls the release time of the outer reservoir, rather than its proximity to the peripheral edge of the silicon wafer. Similarly, D3 is large enough between adjacent reservoirs that it is the “lid” thickness that is corroded first, and not the island walls 74 between adjacent reservoirs (after one reservoir has already been opened to body fluids corrosion of the island walls occurs).
Of course, we may prefer to have the progressive opening of reservoirs achieved by corrosion of the dividing walls between adjacent reservoirs, instead of or in addition to the corrosion of peripheral surfaces of the tablet.
It will be appreciated that the reservoirs of drug material could hold beneficial substance in any form, for example liquid drug, or powder drug, or solid drug. The drug could be a complex organic molecule, or it could be a micronutrient or micromineral as previously discussed.
The reservoirs of drugs could contain a micromineral tablet, or other tablet/implant. A reservoir hole may contain a plurality of erodable drug/element delivery tablets/implants, which may contain the same or different beneficial substances, and/or may be corroded at different/the same rates. Thus, the doors to reservoirs may be separately eroded to allow physiological access to tablets which in turn release a beneficial substance in a controlled manner over days, weeks or months. There may be several tablets in a reservoir, or tens of tablets, or hundreds of tablets.
The “reservoirs” need not necessarily be machined holes in a body of resorbable porous silicon material, they could be regions which have been differentially impregnated with a beneficial substance in comparison with the “walls” (or they could be regions which otherwise have a differential level of beneficial material in comparison with the “wall” regions). Thus, the implant may be a solid body (possibly made from discrete sections, but with no actual holes). However, at present, it is envisaged that micromachining an array of holes will probably be best.
The wall regions can be considered to be time delay means adapted to delay the timing of release of the contents of the reservoirs.
It will be appreciated that silicon technology is indeed ideally suited to compartmentalising drug payloads—an attribute that is exploited in this invention. The basic idea is to micromachine an enormous number (e.g. 102-104) of independent reservoirs into resorbable blocks of Si, thereby generating a new way of controlling kinetics of drug release. The time of release from each reservoir is predetermined by the overlying thickness of a microporous “lid” that is gradually eroded in-vivo.
The example of
The arrangement of
One of the reservoirs in
The multi-reservoir implanted discussed are all fully resorbable, and do not require surgical removal, but the invention is also applicable to non-corrodable implants having corrodable doors. The non-corrodable part of the implant may be bulk silicon.
The above types of delivery system offer much better control and predictability of delivery rates than conventional “monolithic” polymer systems. In the latter case drug release rates are often determined by diffusion through a tortuous pore network, at least for sustained release.
It will also be appreciated that the technical effect achieved by the embodiment of
According to another aspect the invention comprises an implant having a plurality of reservoirs, a plurality of beneficial substance charges provided on said reservoirs, and a plurality of barrier regions, or doors, provided adjacent said reservoirs, the doors having a plurality of different erosion times when implanted, the arrangement being such that in use the doors are broken down sequentially in order to stagger the release of the contents of the reservoirs.
There may be up to ten reservoirs, of the order of tens of reservoirs, or even of the order of hundreds of reservoirs, or more.
The invention also comprises a method of timing the release of beneficial substances in an implant.
The fact that silicon does not resorb too quickly is beneficial. It is preferred to have an implant that will not need replacing for at least one month, and most preferably for at least two months, three months, four months, six months, nine months, or a year or more.
A problem with using the erosion of an implant to deliver drug embedded in the material of the implant is that the surface area of the implant changes with time (or can change with time) and hence the drug delivery changes with time. For example, a sphere gets smaller. This can in part be countered by the geometric design of the implant to allow the creation of an expanding internal surface to compensate for a contracting external surface.
An alternative, or complementary, approach that is now realisable with porous silicon, and with polycrystalline silicon, is to ensure that the drug/beneficial substance is present at different concentrations at different regions of an implant. This can be achieved by controlling the pore size through the depth of a body of porous silicon, or by controlling the grain size/number of grain boundaries. The number and/or size of grain boundaries may be controlled throughout the depth of a body of polycrystalline silicon. Thus, it is possible to have a porous silicon tablet which has a substantially uniform dose delivery rate with time as it is resorbed due to the concentration of drug/micromineral in it increasing towards its centre so as to balance the decrease in exposed surface area.
It will be noted that substantially 2-dimensional shapes, such as a disc, do not suffer so much from changes in surface area, and neither do the elongate “line” shapes (as shown in
In addition to the porosity affecting the amount of substance that can be held in microporous silicon (greater porosity, greater substance-containing capability), the pore size can affect the rate of dissolution of the implant. Thus, the inner regions of a porous silicon implant can be arranged to corrode faster than the outer regions, again having a compensating effect for the loss of exposed surface area.
Whilst many countries do not, yet, permit the patenting of methods of treatment of the human or animal body by surgery or therapy, there are some (e.g. USA) who do. In order for there to be no doubt about the Paris Convention priority entitlement to such an invention in those countries that do permit it, the invention also comprises the treatment, therapeutic or prophylactic, of a disorder of the human or animal body by implanting a silicon implant and allowing the implant to corrode or be resorbed so as to realise a beneficial substance which helps to alleviate or ameliorate the disorder, or to prevent the disorder from occurring. The implant will typically be implanted subcutaneously.
Furthermore, the technique could be used to release diagnostic substances, possibly in a localised region of the body. Diagnostic substances are “beneficial substances”.
It will be appreciated that the realisation that silicon structures, especially porous silicon and polycrystalline silicon structures are able to be broken down by the body over a long (months) period of time without evidence of significant harmful effects has led to the ability to create beneficial substance (e.g. micronutrients and drug) delivery implants which take advantage of this. The evidence showing no detrimental effects of implantation enables us to have a reasonable and predictable expectation of success—it is more than speculation.
At present, it is perceived that restrictions on the physical size of the drug payload for implants will restrict their practical use to delivering microminerals, or other substances, which are not required at high levels (e.g. genetically engineered proteins, peptides, and gene fragments, and other DNA material). However, the invention is not necessarily to be restricted to those areas if a practical macro-drug delivery implant is created.
A “beneficial substance” is something beneficial overall: it could be a toxin toxic to undesirable cells/to interfere with an undesirable physiological process. For example, anti-cancer substances would be considered “beneficial”, even though their aim is to kill cancer cells.
It will be noted that the terms “eroded”, “corroded”, “resorbed” are all used herein. The mechanism of corrosion is not fully known, but that erosion, corrosion takes place is proved. Bioerosion, bioresorption, biodegradation are other possible terms: at present whether the silicon/carrier material is taken up into cells or stays extracellular is not considered important. It is not intended, necessarily, for the invention to be restricted to any precise biological distinction between the “corrosion” terms used.
Impregnation of Porous Semiconducting Materials
Experiments were carried out to demonstrate the impregnation, according to one aspect of the invention, of porous silicon samples with a number of metals (manganese, silver, and chromium) or compounds (for example oxides) of these metals. A salt of the metal was placed on the surface of a sample of porous silicon. The temperature of the salt was raised until the salt melted. The molten salt then passed into the bulk of the porous silicon. The application of heat resulted in decomposition of the salt to yield the metal or the metal oxide.
The starting material was 3-5 ohm cm n-type (100) silicon. This was anodised in a 50/50 mixture by volume of ethanol and 40 wt % hydrofluoric acid. The anodisation current was 100 mAcm−2 and the anodisation time was 5 minutes. This gave a porous silicon film of thickness 30 microns and gravimetrically determined porosity of 38%. Samples of the porous silicon (supported on the un-anodised bulk silicon) prepared by this method, were cleaved to make pieces approximately 2 cm by 2 cm in dimension.
The metal salts chosen were the manganese (II) nitrate, chromium (III) nitrate, and silver (I) nitrate. A general procedure, according to one aspect of the invention, was adopted. The nitrate was placed on the surface of the porous silicon sample. The sample of porous silicon was placed on a graphite block with the porous face upwards. On to the surface was placed an amount of the metal nitrate powder. The graphite block, with the porous silicon wafer on it, was loaded into a CVD reactor. The reactor was assembled and closed to the atmosphere. The CVD reactor was flushed with argon (to create an inert atmosphere) or hydrogen (to create a reducing atmosphere).
The sample temperature was then raised until the metal nitrate was observed to melt.
For some samples, following a period at this initial temperature, the temperature was raised further and the salt observed to decompose by the evolution of bubbles. After some time at elevated temperature the sample was cooled to room temperature and removed from the CVD reactor. A number of the samples produced by this method were then washed in deionised water and dried. After washing, analysis of the metal content was performed by EDX on cleaved edges.
For each impregnation procedure, the sample of porous silicon was weighed prior to impregnation. After impregnation the samples were washed in deionised water and re-weighed. For each of the three nitrate salts tested, an increase in weight was observed. Since the three nitrate salts are all highly soluble in water, the increase in weight suggests that the nitrate salts were decomposed, presumably to either the metal or an oxide of the metal, by the heat applied to the porous silicon sample.
The procedure for managanese (II) nitrate, which is consistent with the above mentioned general procedure, will now be described. Manganese nitrate powder, sufficient to give approximately 0.5 gram of powder per 1 cm2 of porous silicon surface, is placed on the surface of the porous silicon sample. Inert gas (argon) at a rate of 700 cm3 /min was allowed to flow through the CVD reactor for ten minutes. At this point the temperature of the graphite block with the wafer upon it was raised to 50° C. The manganese nitrate was observed to melt and the temperature was maintained at this value (50 C) for one hour. The temperature was then raised to 150° C. and maintained at this value for a further hour. At this stage gas evolution from the molten salt was observed. The temperature was then allowed to cool to room temperature and the sample removed. The sample was then washed by immersion in deionised water for about 5 minutes. This was observed to remove most of the salt remaining on the surface of the porous silicon. Samples for elemental analysis were then cleaved from the sample.
The porous silicon sample, which had been treated with manganese nitrate, was washed in water to remove any excess of the unreacted salt on the surface although a clearly marked area was left which indicated where the salt had melted on to the porous silicon substrate. Elemental analysis on a cleaved section (the cleaving occurring at or close to the boundary between the porous and bulk silicon) through the sample was then used to reveal the extent of the manganese impregnation. In all cases, for the manganese, the metal or more probably the metal oxide had reached the bottom of the porous layer, a distance of 30 μm with the substrates used in these experiments. Manganese was observed only under the area where the molten salt had been. The composition even at the bottom of the pores was sufficient for it to be easily detected by EDX suggesting that it was in excess of one atomic percent. It should be noted that the technique of EDX only allows concentrations of metal in excess of one atomic percent to be detected. For treatment with manganese nitrate, the above procedure was carried out in both a hydrogen atmosphere and in an argon atmosphere. In both types of atmosphere similar concentrations of manganese atoms were observed at similar depths.
The procedure for chromium (II) nitrate, which is consistent with the above mentioned general procedure, was identical to that described for manganese (II) nitrate given above except that the graphite block was heated to 90 C to cause melting and after one hour this temperature was raised to 150 C and maintained at this value for a further hour. The procedure for silver (I) nitrate, which is consistent with the above mentioned general procedure, was identical to that described for manganese (II) nitrate given above except that the graphite block was heated to 250 C to cause melting and after one hour this temperature was raised to 450 C and maintained at this value for a further hour.
EDX analysis for samples of the chromium and silver nitrate treated samples was carried out in a similar way. The treated porous silicon sample was washed in water to remove any excess of the unreacted salt on the surface. Elemental analysis on a cleaved section (the cleaving occurring at or close to the boundary between the porous and bulk silicon) through the sample was then used to reveal the extent of the impregnation of the salt. Chromium oxide (for the chromium nitrate treated sample) had reached the bottom of the porous layer, a distance of 30 μm with the substrates used in these experiments. Silver (for the silver nitrate treated sample) had reached the bottom of the porous layer, a distance of 30 μm with the substrates used in these experiments. Unlike the cases of the manganese and chromium treated samples, silver was distributed throughout the pore structure and not just in the area under the melt. The composition even at the bottom of the pores was sufficient for it to be easily detected by EDX suggesting that it was in excess of one atomic percent.
The impregnation procedure for manganese was also carried out in ambient air. Manganese nitrate was placed on the surface of a porous silicon film; the film being placed on a standard laboratory hotplate. The sample was heated on a hotplate to 56 C for 70 minutes and 150 C for 70 minutes. This gave a black film on the surface of the porous silicon layer. EDX analysis of this film revealed manganese at greater than 1% throughout the layer. There was also a band of higher concentration (a few atomic percent) at a depth of a few microns.
Similar methods to that described here could be used to pass impregnate substances other than metal salts into any porous semiconductor (porous silicon materials being a subset of porous semiconductors). The impregnate substance could be a metal salt (including the metal nitrates described here) and/or a beneficial substance. The impregnate substance could be an element of the periodic table. Samples of porous silicon, having a sample surface and impregnated by identical or similar methods to those described here, could be used as a component in implants described at other parts of this application.
Invivo Trial of Porous Silicon Implant
The purpose of the trial was to evaluate the biocompatibility of porous silicon when implanted at subcutaneous sites in guinea pigs in order to investigate the materials' suitability for use in implantable devices. The trials were carried out over 1, 4, 12, and 26 weeks.
Experiments were conducted in accordance with the methods specified in the International Standard for biological evaluation of medical devices part 6 (ISO 10993-6).
The test specimens were in the form of discs (10 mm in diameter, 0.3 mm in thickness). Manipulation of their surface characteristics aimed to make 1 specimen type bioactive (ie encourage tissue attachment; hereafter termed porous silicon), 1 specimine type bioinert (ie produce no interaction with living tissues; hereafter termed bulk silicon) and 1 specimen type bioactive pre-coated with hydroxyapatite (hereinafter termed coated porous silicon). One of each specimen type and one control (titanium disc of the same dimensions as the test specimens) were used per animal in the 1, 4, and 12 week study. Two porous silicon samples and two titanium samples were used per animal for the 26 week study.
The 1, 4, and 12 week trials used a total of 30 guinea pigs (10 guinea pigs for each time period). The 26 week trial used a further 5 guinea pigs, making a total of 35 guinea pigs. There was a pilot phase of the trial for seven days prior to the 1, 4, 12, and 26 week periods. The pilot study was carried out on three guinea pigs (one from each of the 1. 4. and 12 week groups). The pilot study was successful (ie no gross reactions to the implants occurred) so the full scale study proceeded as planned.
Animals were acclimatised to the Experimental Animal House (EAH) environment for at least 5 days prior to experimentation. Following this period each animal was implanted with a transponder (Biomedic Data Systems) for identification and to enable body temperature to be monitored throughout the procedure. The transponder was implanted subcutaneously via a 12 gauge needle in the dorsal region, at a site where it did not interfere with the subsequent implantation of silicon or control specimens. The area of injection was shaved and a local anaesthetic used.
4-7 days later animals were given a general anaesthetic (Halothane 1.5-2.5%) and 4 test specimens implanted. The back of the animal was shaved and an incision of the skin made. Subcutaneous pockets were made by blunt dissection, with the base of the pocket at least 15 mm from the line of the incision. One implant was placed in each pocket, and the implants were at least 5 mm from each other. Four pockets were made to allow placement of 4 implants. The incision was closed by use of appropriate suture material.
Body temperature (via the transponder) was measured twice a day following surgery for the duration of the study (including the seven day pilot study). Each of the implant sites was closely examined and the extent of any reaction noted. The diameter of each implant site was measured to assess swelling and any reddening scored (0=normal, no different from surround skin; 1=some light red coloration in patches; 2=uniform light red coloration or patches of darker red; 3 darker red over all of implant site). At the end of the relevant study period (1, 4, 12 or 26 weeks) animals were killed by an overdose of pentobarbitone. The implant sites were carefully inspected and standard tissue sections of each site were submitted, stained with haematoxylin and eosin and evaluated for various pathological features using a Zeiss Axioplan Photomicroscope. A range of pathological features reflecting the tissue response including acute inflammation and fibrosis were graded by assigning a numerical score to each feature; by comparing score grades with respect to time and implant site, an objective comparison of the silicon materials was obtained. The criteria used in assigning score grades for each pathological feature assessed are summarised in Tables A to D. The specimen type at each implant site was randomised and the experiments and evaluation conducted blind.
The scores or values for each specimen type and each time point were compared with those of the control specimens using appropriate non-parametric tests. Multi-factorial analysis of variance was used where possible, with ad hoc tests for differences between groups.
Mean temperature and weight data is shown in graphical form are shown in
At the time of performing the histological appraisal, the allocation of test and control sites for each experimental animal was unknown and the histological examination was performed blind. Following appraisal, the results were decoded: a summary of the score grades for each implant type with respect to animal number, histological feature and timepoint are summarised in the tables E to H. In general the autopsy examination revealed no evidence of any significant pathological change at any of the three time-points. In particular, all implants were easily extracted from their respective implantation sites and showed no evidence of fibrotic tethering to the surrounding connective tissues. At the earliest time-point each site showed obvious acute inflammation associated with mild to moderate neo-vascularization. At 26 weeks, three out of the twenty sites examined showed mild to moderate chronic inflammation/fibrosis around the vicinity of the implantation site consisting of a cuff of macrophages, lymphocytes and occasional foreign body giant cells. In each casethese changes were almost exclusively limited to the implantation site. The histological findings were entirely consistent with the features noted at autopsy. The scores for each of the four classes of pathology (Tables E to H) were compared with respect to time (ie week 1 vs. week 4 vs. week 12 vs week 26) and implant type (the scores of each silicon type compared with the titanium control). Details of the statistical analysis are shown at Tables I and J.
Acute inflammation at week 1 was significantly greater than at weeks 4 and 12 but no difference was found between weeks 4, 12, and 26 (Table I).
Tissue degeneration was significantly higher at weeks 1 and 4 when compared to week 12, with no difference between weeks 1 and 4. There was no significant difference in tissue degeneration/necrosis between test and control samples in any of the weeks. New vessel/granulation tissue formation was significantly higher at weeks 1 and 4 than at weeks 12 and 26; there was no significant difference between weeks 1 and 4. Chronic inflammation was significantly higher at week 4 than at weeks 1, 12 and 26; and it was significantly higher at week 12 than week 1. In general, these significant findings were consistent with the three distinct patterns of pathological change observed at the three excision time-points, summarised below.
All sites at one week post-implantation showed features that reflected the immediate response to the injury induced by the surgical procedures to implant the materials. Most sites showed moderate acute inflammation with infiltration of the tissues at the implantation site by neutrophils and macrophages. These changes were associated at the majority sites with oedema of the connective tissues, focal haemorrhage and necrosis and early invasion of the margins of the implant site by proliferating capillary loops. At no site did these changes extend beyond the margins of the implant and into the surrounding skin above or skeletal muscle below.
Although a very few sites showed persistent, low grade acute inflammation four weeks post-implantation, the majority of sites showed features that were consistent with the progression of the features described at one week and represented attempts at tissue repair following surgery rather than a reaction to the silicon implant. Most sites showed areas haemorrhage surrounded by loose granulation tissue, active proliferation of new blood vessels and a limited population of active fibroblasts. In a few cases these reparative features extended beyond the implantation site but, even in these cases, did not cause major disruption to the surrounding tissue architecture.
By twelve weeks, the histological features represented a maturation of the granulation (repair) tissue response observed at four weeks. Although many of the implantation sites still showed significant infiltration by macrophages, lymphocytes and occasional fibroblasts, they showed no evidence of significant fibrotic scarring and a definite reduction in vascular proliferation. Furthermore, in no case did the persistent pathological change extend beyond the implantation site.
In general after 26 weeks, the implantation sites all showed little evidence of any significant tissue reaction to either the test or standard implants. The sites displaying mild to moderate chronic inflammation around the immediate vicinity of the implantation site consisted of a cuff of macrophages, lymphocytes and occasional foreign body giant cells which did not involve the continuous soft and connective tissues and were not associated with distorting fibrosis of nearby structures.
The major internal organs were also examined at autopsy after 26 weeks implantation, with representative blocks being submitted for routine histopathological examination. In keeping with the observations made at the time of autopsy, histological examination of the major internal organs revealed no evidence of any pathology that could be ascribed to the silicon or titanium implants or any pre-existing disease in the experimental population.
The post-evaluation analysis of the scores for each implant type revealed a significantly higher level of chronic inflammation/fibrosis at weeks 4 and 12 for the porous silicon (uncoated) specimens when compared to the titanium controls (Table J). The nature of the tissue reaction noted is likely to be a reflection of the bioactive nature of the porous silicon implant type, suggesting that this material encourages tissue growth and interacts with biological systems. No other statistically significant differences were revealed for the other histological features or implant types at any of the time points.
The results of this study clearly demonstrate that there has been little or no reaction either the test or standard implant materials. The significant differences in histological features reflect changes which would be expected from any surgical procedure and are unrelated to the nature of the implant materials.
The significant differences in the chronic inflammation scores for the porous silicon at weeks 4 and 12 highlighted by the multivariant analysis are unlikely to be biologically significant in terms of biocompatability. This conclusion is confirmed by the results of the 26 week study.
Score Grades
Tables A to D indicate the score grade criteria that were used to assess acute inflammatory reaction; tissue degeneration, oedema formation, haemorrhage and skin necrosis; new vessel and granulation tissue formation; and persistent (chronic) inflammation and tissue fibrosis during pathology.
Tables E to H show the pathology score grades for 1, 4, 12, and 26 weeks after implantation, respectively.
Table I shows a statistical analysis of the biocompatibility study. For table I average rank of scores for each histology category for each time period. In table I a line between two rows in the significance column indicates a significant difference for those two groups (p<0.05; Kruskall-Wallis analysis).
Table J shows statistical analysis of the biocompatibility study. For table J average rank of scores for each implant type by histology category for each time period. In table J “*” denotes a significant difference between the rank for that silicon type in comparison to the titanium control (p<0.05; Friedman analysis), where BSi=Bulk silicon, PSi=Porous silicon, CoPSi=Coated porous silicon.
| Number | Date | Country | Kind |
|---|---|---|---|
| 9808052.6 | Apr 1998 | GB | national |
This application is a divisional of application Ser. No. 12/801,825 filed Jun. 28, 2010, which in turn is a divisional of application Ser. No. 09/647,599 filed Oct. 3, 2000, now U.S. Pat. No. 7,763,277, which in turn is a US national phase of PCT application Serial No. PCT/GB99/01185 filed Apr. 16, 1999 which in turn claims priority of GB Serial No. 9808052.6 filed Apr. 17, 1998; the entire contents of each of these applications is hereby incorporated by reference in this application.
| Number | Date | Country | |
|---|---|---|---|
| Parent | 12801825 | Jun 2010 | US |
| Child | 12805778 | US | |
| Parent | 09647599 | Oct 2000 | US |
| Child | 12801825 | US |
