EP0741581A1

EP0741581A1 - Oral magnetic particle formulation

Info

Publication number: EP0741581A1
Application number: EP95906454A
Authority: EP
Inventors: William George Gorman; Fred Andrew Carroll; Kevin Darryl Ostrander; William Hunke
Original assignee: Nycomed Imaging AS
Current assignee: GE Healthcare AS
Priority date: 1994-01-28
Filing date: 1995-01-30
Publication date: 1996-11-13
Also published as: CN1142191A; AU1464595A; MX9603001A; JPH09510698A; CA2182213A1; WO1995020405A1

Abstract

There is provided a substantially hydrated, low viscosity carrier for dispersing magnetically responsive particles. Contrast media for nuclear magnetic imaging of the gastrointestinal system are also provided for oral or rectal administration. A method of generating a magnetic resonance image using the composition of the invention is also described.

Description

ORAL MAGNETIC PARTICLE FORMULATION

The present invention relates to magnetic resonance imaging (MRI) in general and in particular to compositions useful as or in the preparation of MRI contrast media for imaging the gastrointestinal system or ^" other body cavities from which contrast media may be discharged without passing through body tissue.

The use of contrast agents for effective MRI of the gastrointestinal tract is well established. Both positive and negative contrast agents have been described, the latter agents generally containing ferromagnetic, ferrimagnetic or superparamagnetic particles. These particles are usually dispersed in a liquid carrier to form a suspension which is administered to the site of interest for imaging.

Among negative contrast agents, magnetically susceptible iron oxide (MSIO) particles are attractive because they are miscible with intestinal contents and are effective at small doses. MSIO agents are superparamagnetic and ferromagnetic agents which produce signal blackening (negative contrast) by creating a local distortion of the magnetic field, resulting in T2 relaxation time shortening.

Moderate magnetic field distortion causes signal blackening, but excessive magnetic distortion can result in poor imaging by causing magnetic susceptibility artifacts (pixel mismapping and image distortion) . Excessive magnetic distortion is usually caused by high particle concentration, but even at a suitable concentration, excessive magnetic distortion can also be caused by particle aggregation and flocculation, frequently seen vJ.th suspensions (such as MSIO suspensions) that are susceptible to gravitational settling. Since magnetic susceptibility artifacts cause distracting bright signals, interfere with visualization of normal adjacent structures and induce significant image blurring, they have limited the widespread acceptance of MSIO contrast agents. Accordingly, although several MSIO contrast agents have recently been developed, they have not been widely used.

In general, satisfactory distribution of the contrast medium in the bowel loops is difficult to achieve due to both the meandering configuration and tremendous surface area of the region. Additionally, peristaltic movements can further affect distribution. To overcome this problem, practitioners can use high concentrations of magnetically responsive particles sufficient to produce the necessary contrast effect throughout the imaged zone, but this can cause the local concentration of particles in the region to become so high that image distortions akin to "metal artifacts" are produced. This is highly undesirable as such artifacts might be mistaken for pathological structures and since the most important function of the contrast medium in such imaging is to allow reliable differentiation between the body cavity containing the contrast medium and pathological structures in the body, particularly in the abdomen, any such uncertainty seriously reduces the diagnostic value of the technique.

In short, while magnetically responsive particles are extremely effective in enhancing image contrast, several workers have concluded that negative contrast media are of little value or of less value than positive contrast media in abdominal imaging.

Attempts have been made to formulate a product with sufficient viscosity to maintain particle dispersion and still avoid particle aggregation and flocculation. The problem is that in order to achieve good images, the gastrointestinal tract must be maximally filled which means that the patient must consume large volumes of the bulky formulation. While dispersion of the particles in the formulation is improved if its viscosity is high, ingestion or infusion of large volumes of highly viscous contrast media is difficult or impossible for the patient.

To overcome this problem, Nycomed (PCT/EP90/01196) discloses a negative contrast medium which contains viscosity enhancing agents that reach full viscosity enhancing effect only after administration . In the Nycomed formulation the vi scosity enhancing agent is " incompletely hydrated" , meaning that it reaches full viscosity only after exposure to aqueous media, such as water or body fluids like gastric juices .

The problem with that approach is that if the product is incompletely hydrated, the active ingredient is not properly distributed to yield superior MR-images upon ingestion . Furthermore there is no proof that post- ingest ion distribution/hydration is complete or optimal, again jeopardizing MR- image efficacy. In addition, preparation (reconstitution) of granular product just prior to dosing is subject to human error which., may result in improper concentration and thus may adversely affect image quality.

What is needed, then, is the development of a ready-to- use formulation which ensures complete hydration and consistent dispersion of the active particles prior to dosing offers superior MR-image efficacy and overcomes potential errors in preparation .

Another problem encountered with oral magnetic particle (OMP) preparations is that on storage or exposure to heat they rapidly lose viscosity and develop caking, thereby losing imaging efficacy.

The paradigm for an efficient OMP is a good-imaging, low-viscosity, fully hydrated, ready-to-use formulation that is capable of uniformly dispersing the magnetic particles without being so viscous that it is unpalatable . This formulation would contain an appropriate dispersing agent in a sufficient amount . If it is to be stored for long periods prior to administration, a useful feature, it must be preserved against microbial activity and must resist caking that often occurs when active ingredients and excipients settle and adhere to the walls of containers . For better patient compliance, it should preferably be as palatable as possible which means, at the least, that the formulation should mask the taste and color of the act ive metal lic ingredient . This would involve selecting appropriate sweeteners and coloring agents which do not react with the metal particles . - 4 -

It would be highly desirable to obtain such a formulation.

Prior to this invention, attempts were made to overcome the problem of imaging artifacts and poor images obtained with simple (i.e. without thickeners) aqueous suspensions of magnetically responsive particles. These efforts focused on increasing the viscosity of the suspension.

We have found that to obtain high quality images, the important feature of the suspension is not increased viscosity, but rather, the ability of the carrier to maintain dispersion of magnetically responsive particles to form a stable homogenous suspension. Viscosity is important only to the extent that it prevents agglomeration or gravitational settling of particles and affords the formation and stability of a homogeneous suspension.

Accordingly, the present invention provides formulations for contrast media with viscosities as low as 25 to 465 cP which produce high quality images. These formulations suspend magnetic particles very well for extensive periods of time without caking, lumping or separation. Further, the product is substantially hydrated at the time of consumption.

Thus, in one aspect of the invention there is provided an aqueous carrier for dispersing magnetically responsive particles, the carrier comprising one or more substantially hydrated dispersion-enhancing agent.

The invention also provides a contrast medium comprising a suspension of magnetically responsive particles dispersed in a substantially hydrated carrier.

Yet another aspect of the invention provides a ready-to- use contrast medium which remains'unchanged after storage.

A method of generating a magnetic resonance image of a human or nonhuman body is also provided. O 95/20405

-5-

Figure 1 represents four plots showing the relationship between viscosity and signal intensity ratios.

Figure 1A represents the-Tχ-weighted pulse sequence.

Figure IB represents the T₂~weighted pulse sequence.

Figure 1C represents the proton density-weighted pulse sequence.

Figure ID represents all three pulse sequences combined.

Figure 2 is a plot showing the viscosity values for various formulations of the invention measured during the course of study.

As used herein, the following terms shall have the following meanings.

"Substantially hydrated" means that one or more dispersion enhancement agents in the formulation are in contact with water, such that a desirable viscosity is achieved prior to patient dosing. A desirable viscosity is one that is palatable, yet is capable of uniformly dispersing the magnetic particles. While it is possible that viscosity might increase during passage to the site of interest, such increase is not critical for high performance.

"Magnetically responsive particles" relates to particles having paramagnetic, diamagnetic, and especially ferromagnetic, ferrimagnetic or superparamagnetic properties. Such particles produce blackening by creating local distortion of the magnetic field, resulting in T2 relaxation time shortening. agnetically susceptible iron oxide (MSIO) is exemplary.

"Ready-to-use" means that the product is capable of being administered to the patient directly from the package with no preparation other than mild shaking. No diluting or reconstitution is required. - 6 -

"OMP" means .oral magnetic particles , indicating that magnetically responsive particles have been added to a liquid carrier to provide a suspension for oral consumption.

"Carrier" means a chemical vehicle used to assist or transport the active component to the desired anatomical site.

"Phantom" refers to a twin coaxial tube apparatus of which the outer tube contains water and the inner tube contains MRI contrast medium. It is used for in-vitro MR evaluation.

"Negative contrast agent" describes an agent which contains materials whose effect of reducing the spin-spin relaxation time (T2) of the imaging nuclei outweighs any Tl reducing effect and results in a reduction in MR signal intensity from the body regions into which they distribute . Negative contrast media generally contain ferromagnetic, ferrimagnetic or superparamagnetic particles .

"Positive contrast agent" describes a paramagnetic compound which shorten the spin-lattice relaxation time (Tl) of the imaging nuclei and so result in an increase in image intensity in the body regions into which they distribute . One such positive contrast agent is Gd DTPA.

"LVL" denotes lowest yiscosity limit. Formulations with viscosities lower than the LVL are prone to gravitational settling.

The present invention provides a fully hydrated, low viscosity carrier which contains one or more dispersing agents in sufficient amounts to uniformly disperse magnetically responsive particles . When suitable magnetically responsive particles are suspended in the carrier, the composition is a contrast medium which affords high quality magnetic resonance images .

Suitable dispersing agents are capable of dispersing magnetically responsive particles under the physiological conditions of the body cavity to be imaged and thus are preferably of a non-biodegradable material, especially where the composition is intended for oral ingestion. Dispersing agents may conveniently be soluble in aqueous media to produce a viscous solution . Examples of such materials include natural , semisynthetic and synthetic high molecular weight substances such as natural or semi synthetic gums and polysaccharides, e . g. guar gum, tragacanth , methylcellulose , hydroxypropylcel lulose, carboxymethylcellulose, xanthan gum, a lginates and, where applicable, their physiologically acceptable salts . Many examples of such materials are known as thickening agents in the food industry .

Alternative dispersion enhancing agents include insoluble materials which swell in aqueous media to produce viscous dispersions . Typical examples of such swellable dispersion enhancing agents include clays, eg kaolin, and related minerals such as, for example, magnesium aluminum silicate, bentonite, etc . Mixtures of soluble and insoluble dispersion enhancing agents can also be used.

For administration into the GI tract, bulking agents such as those used in the treatment of constipation such as bran, psyllium and methylcellulose may also be used as dispersion enhancing agents alone or in combination with other dispersion enhancing agents .

Preferred dispersing agents include carboxymethylcellu¬ lose sodium, hydroxypropylmethylcellulose, methylcellulose, hydroxyethylcellulose, hydroxypropylcellulose, microcrystalline cellulose, carbomers, gum tragacanth, sodium alginate, gelatin, pectin, polyvinylpyrrolidone, guar gum, xanthan gum, pregelatinized starch, locust bean gum, montmorillonite, bentonite, hectorite, carrageenan, starch, xylitol, sorbitol, mannitol, and lactose.

The complete contrast agent of the invention comprises magnetically responsive particles, the components actually responsible for the negative MRI contrast , dispersed in the carrier composition of the invention.

As mentioned above, many forms of magnet ically responsive particles have been proposed for use as MRI contrast agents and generally speaking, all such particles may be used in the carrier composition of the invention . Thus the particles may be free or may be coated by or embedded in or on particles of a non-magnetic carrier material , e . g . a natural or synthet ic polymer, for example cellulose or a sulphonated styrene-divinyl benzene copolymer (see for example WO83/03920 of Ugelstad) . The magnetically responsive particles may be ferromagnetic or ferrimagnetic or may be sufficiently small as to be superparamagnetic and indeed superparamagnetic particles are generally preferred.

Thus, the magnetically responsive particles used according to the present invention may be of any material which (although preferably non-radioactive unless the particles are also intended to be detected by their radioactive decay emissions) exhibits ferromagnetism, ferrimagnetism or superpara- magnetism. The particles may conveniently be particles of a magnetic metal or alloy, eg of pure iron, but particularly preferably will be of a magnetic compound such as a ferrite, for example magnetite, gamma ferric oxide and cobalt, nickel or manganese ferrites.

To avoid image distortion, it is preferred that the mean particle size of the magnetically responsive particles be less than about 5 micrometers, preferably less than 1 micrometer and that the overall size of the non-magnetic carrier particles be less than 50 micrometers, preferably less than 20 micrometers, especially preferably 0.1 to 5 micrometers. The magnetically responsive particles will generally have mean particle sizes in the range 0.002 to 1 micrometers, preferably 0.005 to 0.2 micrometers.

Where the magnetically responsive particles are carried by carrier particles, these are preferably of a material which is physiologically tolerable and which is not biodegradable, at least in the environments it will experience on the way to and at the body cavity being imaged.

The compositions of the invention may, include components other than the dispersion enhancing agent and the magnetic particles. For example, conventional pharmaceutical formulation aids such as wetting agents, disintegrants, binders, fillers, dyes, osmoactive agents, flavoring agents and liquid carrier media. To improve contact between the magnetically responsive particles and the walls of the body cavity, e.g. the gut wall, the compositions may also contain mucoadhesives, such as for example a polyacrylic acid or a derivative thereof, xanthan gum, etc.

The compositions of the invention are particularly suited for use as MRI contrast media for imaging of the gastrointestinal tract and in particular for imaging the duodenum and the intestines. For such purposes the contrast medium may be administered orally or rectally or by orally or rectally inserted tubes. However, as indicated above the contrast media are also suitable for use in imaging other externally voided body cavities such as the bladder, uterus and vagina.

When the composition is to be stored, as in the case of ready-to-use preparations, package compatibility studies indicate that product containers composed of glass or polyethylene terepthalate (PET) afford the greatest product shelflife with regards to preservation efficacy.

Thus, viewed from another aspect the present invention provides the use of a physiologically tolerable dispersion enhancing carrier for dispersing magnetic particles in a composition. The composition is suitable for use in magnetic resonance imaging.

Viewed from a further aspect, the present invention provides a method of generating a magnetic resonance image of a human or non-human, e.g. mammalian, subject in which method a contrast medium comprising magnetically responsive particles in a dispersion enhancing carrier is administered into an externally voided body cavity of the subject (e.g. the gastrointestinal tract) , wherein said dispersion enhancing carrier acts to increase the dispersion of the magnetically responsive particles following administration of said medium into the subject.

Viewed from a yet further aspect, the present invention provides a packaged, substantially hydrated, ready-to-use diagnostic contrast agent comprising a plurality of magnetically responsive particles dispersed in a physiologically tolerable dispersion enhancing agent.

In the method of the invention the dose of the contrast medium will generally be at least 500 mL for an adult human subject and more usually 600 to 1100 mL, especially 750 to 1000 -10-

ml. The content of the magnetically responsive particles will depend on the particular particles used. However, the particles will generally be contained at a concentration of 0.01 to 10 g/litre, preferably 0.1 to 3 g/litre. The dose may be taken in portions, e.g. for oral administration about 2/3 being ingested 20 minutes before imaging and the remainder being ingested immediately before the subject is placed within the magnet (or scanner) .

The formulations for the dispersion enhancing carrier and the contrast medium are shown in Table I.

Table I

All units listed are to be taken as w/v, except * and **.

* μg Fe/mL

** pH units

*** Avicel CL-611 contains microcrystalline cellulose and carboxymethylcellulose sodium. The invention is further illustrated by the following non-limiting examples.

Example 1 Preparation of 1 Liter of OMP Carrier

Preparation of Dispersion No. 1:

To 750 g of purified water heated to 70°C - 80°C in a jacketed stainless steel or glass-lined manufacturing vessel, the microcrystalline cellulose and carboxymethylcellulose sodium were added and dispersed with vigorous mixing and side scraper agitation. With continuous mixing xanthan gum was slowly added until it was wetted and completely dispersed. Temperature was maintained at not less than 70°C. The dispersion was then cooled to 25°C - 35°C with continued side scraper agitation.

Preparation of Slurry No.l;

To approximately 25 g of purified water in a stainless steel or glass-lined manufacturing vessel, the FD&C Red No. 40, FD&C Yellow No. 6, and FD&C Blue No. 1 were added and dispersed.

Preparation of Solution No. 1;

To approximately 125 g of purified water in a stainless steel or glass-lined manufacturing vessel, the potassium sorbate, sodium benzoate; and saccharin sodium were added and dissolved. Slurry No. 1 above was then added and mixed until completely dispersed.

Preparation of Final Product:

Solution No. 1 immediately above, was then added to Dispersion No. 1 above, and mixed for 15 minutes, and then brought to 95% of 1 liter (the desired final weight (volume) ) with purified water and again mixed for not less than 10 minutes. - 13-

The pH of the dispersion thus obtained was adjusted within the pH range of 4.2 - 4.3, using IN Hydrochloric Acid and/or IN Sodium Hydroxide and again mixed for not less than 10 minutes to ensure uniformity. The dispersion viscosity was measured to ensure a viscosity of 395 cP or higher. When necessary to increase the viscosity, an additional 10% of the formula amount of Xanthan Gum was slowly added with continuous mixing for not less than 15 minutes. The dispersion was then screened throught a 100 mesh stainless steel sieve, into a tared stainless steel or glass- lined mixing vessel.

Example 2

Preparation of Contrast Agent

A homogenous suspension of magnetically responsive particles was prepared by inversion mixing of the active ingredient (MSIO from Nycomed) for not less than 1 hour until it was confirmed that all solids were suspended. A calculated amount of MSIO to achieve a total iron concentration of 150 mcgFe/mL was slowly added with stirring to the OMP carrier described in Example 1 hereinabove. Stirring continued for not less than 15 minutes. The weight (volume) of the final suspension was determined and if necessary adjusted to the final weight (volume) (1 liter) with purified water. Slow mixing continued for not less than 10 minutes to ensure uniformity. The final product was adjusted with IN Hydrochloric Acid and/or IN Sodium Hydroxide to a pH range of 4.2 - 4.3.

Evaluation of Prepared Contrast Aoents

In Vitro and In Vivo evaluations of the contrast agent of the invention were carried out as follows.

Contrast Agents

The active ingredient in the contrast media used in the examples is MSIO (supplied by Nycomed AS, Oslo, Norway) consisting of monodisperse 3-4 μm polymer particles coated with approxi ately 50 nm particles of iron ferrite. (Described in Norwegian Patents 142022; 143403; and 155316) .

Three formulations shown in Table 2 and two formulations in Table 3 were prepared with an iron concentration of Nycomed's MSIO at 125-150 μg/mL, and a viscosity of 156 cP to 430 cP. Xanthum gum was used to control" product viscosities.

The L-type formulations (LI, L2, L3) shown in Table 2 are the ready-to-use preparations of the invention described hereinabove. The G-type formulations (Gl and G2) shown in Table 3 differ from the L-type in composition and, further, are prepared from a granular concentrate reconstituted with water just prior to administration. The G-type preparation represents the prior art.

Animal Preparation

The animals -for in vivo studies were mongrel dogs anaesthetized with 2% Surital™ I.V. (Thiamylal Sodium by Parke Davis) titrated to maintain light anaesthesia. To deliver OMP, a nasogastric tube was inserted into the stomach of the animals and they were positioned in the MR scanner in the supine position.

MR I aσinσ

In Vitro Evaluation: In vitro MR imaging, which was obtained using a Signa (GE Medical Systems , Milwaukee) superconducting magnet system operating at 1.5T, was performed on phantoms containing 20 mL of WIN 39996 samples immediately after manual shaking. A head coil and conventional 2DFT spin-warp technique were used . Multi-section axial acquisitions were obtained with a 24 cm field of view, 256x256 imaging matrix, 7 mm section thickness, and 3 mm intersecton gap . Pulse sequences used were Tl-weighted (TR/TE 300/15 msec, 1 excitation) , T2- weighted (2000/70, 1) , and proton density-weighted (2000/20, 1 ) .

In Vivo Evaluation : In vivo MR imaging, which was obtained using the same MR unit described above, was performed using the head coil and conventional two dimensional Fourier - 1 5-

Transform (2DFT) spin-warp technique. Multi-section axial acquisitions on the abdomen were obtained with a 25-30 cm field of view, 128x256 imaging matrix, 7 mm section thickness, and 3 mm intersection gap. Pulse sequences used were Tl-weighted (TR/TE 300/15 msec, 4 excitations), T2-weighted (2000/70, 1), and proton density-weighted (2000/20, 1) .

Viscosity Measurement

All viscosity measurements were made at 25°C using a Brookfield RVTDV-II Viscometer operating at 10 rpm with a No. 2 spindle. All viscosity values are reported in centipoise (cP) .

Statistical Analysis

The influences of viscosity on imaging qualities were studied by examining the "Signal Intensity ratios vs. Viscosity" plot (Figure 1) . Data were obtained from formulations LI and L3, respectively, regardless of their storage times and temperatures. The relationship was biphasic on each of the three pulse sequences and on all three pulse sequences combined. Both segments of each curve are linearly related. Since this plot indicated a biphasic relationship, the data were analyzed by segmented linear regression analysis. The critical point (the interception between the two regressions) was derived based on the slope and Y-intercept of each regression.

MR Imaging and Imaσe Analysis

The in vitro and in vivo MR imaging procedures using T1-, T2- and proton density-weighted pulse sequences, have been described above. MR images from this study were evaluated both qualitatively and quantitatively. Qualitative assessments involved evaluation of image quality along with image artifacts in terms of their presence or absence, as well as the severity and quantity of bright magnetic susceptibility image distortion, as compared to unenhanced images. Quantitative assessments involved measurements of signal intensity (SI) values from the phantoms for in vitro studies, and from two regions of interest (ROI) in the stomach and four ROI's in the small bowel for in vivo studies. The SI values were normalized by means of the SI ratio: the SI of the contrast agent of the invention over that of air. Ratios less than 2.0 were considered to be "sufficiently black", greater than 2.0 "insufficiently black", and close to 2 "marginal". This criterion was chosen based on preliminary results indicating that ratios above 2 tended to produce insufficient blackening along with image distortions. This criterion was supported by the results of the present studies Table 4.

If? ViVQ Determination of Lowest Viscosity Limit for Excellent

Image Quality

This experiment demonstrates the lowest viscosity that the contrast agent of the invention must maintain to ensure efficient MR imaging.

About 500 - 600 mL of six liquid formulations of varying viscosities targeted to be (1, 25, 50, 75, 100 and 150 cPs designated formulations A,B,C,D,E and F respectively) were prepared and administered within 5 minutes via a nasogastric tube into the stomach of two dogs.

Images of the sites of interest were obtained and evaluated as described hereinabove.

The viscosities of all formulations were very stable as is reflected by the small fluctuations in the viscosity values (Standard Deviation or SD < 3.5 cP) measured during the course of the study. (Figure 2 and Table 4) .

In the in vivo imaging evaluation, excellent signal blackening without image artifacts was produced at viscosities _>~25 cP, with the exception of imaging the small bowel using the proton-density pulse sequence with the 100 cP formulation (Table 4) . The unexpectedly poor quality produced by the 100 cP formulation in imaging the small bowel was caused by a "ghost artifact", which was caused by motion, especially respiratory _ - _γ_

motion. The "ghost artifact" was confirmed by an extra signal overlapping onto the real signal across the entire image. On all three pulse sequences, the 25 cP formulation produced excellent signal blackening with mild image blurring. This mild image artifact was not considered to be significant. This demonstrates that in anaesthetized dogs, contrast media of the invention having a viscosity of at least of ~25 cP produces excellent signal blackening without significant image artifacts.

Table 2 L-type Formulations used in Comparative Study

Formulation

Ingredients (% w/v)

LI L2 L3

Magnetic Particles 0.47% 0.47% 0.47%

HP-Methylcellulose • • 1.27xxx ..

Microcrystalline Cellulose and Carboxymethylcellulose Na 0.480xx 0.250xx 0.280xx

Xanthan Gum 0.120xx 0.0700x 0.180xx

Potassium Sorbate O.lSOxx 0.150xx 0.150xx

Sodium Benzoate 0.120xx 0.120xx 0.120XX

Saccharin Sodium 0.0200X 0.0200X 0.0200x

Sodium Sulfate, Anhydrous • . 0.140xx 0.0140xx

FD&C Yellow No. 6 0.0100x OOlOOx OOlOOx

FD&C Red No. 40 0.00900 0.00900 0.00900

FD&C Blue No. 1 0.00100 0.00100 0.00100

HCl/NaOH q.s. pH-4.25 pH-4.25 pH-4.25

1 Purified Water q.s. (mL) 100.XXXXX lOO.xxxxx lOO.xxxxx

To provide total iron concentration of "150 μg/mL.

-1 9-

In Vi tro Determination of Lowest Viscosity Limit for Excellent

Image Quality

Whereas in vivo LVL was found to be -25 cP, the in vitro LVL was found to be between -25 and -50 cP (Table 4) . The exact reasons for the discrepancy between the in vivo and in vitro LVL values are unknown. It may be the result of in vivo peristaltic activity continuously mixing intraluminal contrast agent and preventing gravitational settling. It may also be the result of enhancement of signal blackening by intraluminal contrast/medium that was above and below the plane of image. It appears that the imaging quality of the contrast agent is dependent on the degree of dispersion of the magnetically susceptible iron ferrite in the formulation, and that a minimum viscosity is needed to insure such dispersion. Formulations with viscosity lower than the LVL value are prone to gravitational settling, resulting in particle aggregation which in turn leads to uneven distribution of particle concentrations. Inhomogeneity in particle concentrations in turn causes insufficient signal blackening and magnetic susceptibility artifacts.

Comparative Formulations

This example compares the imaging qualities of five different OMP formulations (See Tables 2 and 3) after these formulations had been exposed to various temperatures for various periods (Table 5) .

Three liquid formulations (LI, L2 and L3) , which differed in hydroxylpropyl-methylcellulose, xanthan gum, sodium sulfate, microcrystalline cellulose and carboxymethylcellulose sodium concentrations (Table 2) , were studied in vi tro.

Formulation L2 was inferior to formulations LI and L3 because after one-week exposure to 70°C, poor signal blackening

(reflected by SI ratios greater than 2) .on Tl-weighted and proton density-weighted sequences were produced by formulation L2 (Table

5) . Conversely, excellent signal blackening (reflected by SI -20-

ratios smaller than 2) on all three pulse sequences were produced by formulations LI and L3 after they had been exposed to the identical conditions. Formulation L2 was judged to be inferior also because although excellent signal blackening (reflected by SI ratios less than 2) were observed on all three pulse sequences after had been exposed to 30°C for one week, poor signal blackening on the second Tl-weighted sequence was observed. This suggests that gravitational settling occurred 20 minutes after manual shaking. Gravitational settling was not apparent 20 minutes after manual shaking with formulations LI and L3 after these formulations had been exposed to 30°C for one month and one week, respectively (Table 5) .

For the two granular formulations (Gl and G2) examined in this study (Table 3), neither was found acceptable. Poor signal blackening was produced by formulation Gl after exposure to 50°C for six weeks, and by formulation G2 after exposure to 50 and 70°C for one week (Table 5) . For formulation Gl, the efficacy was slightly improved by using glass instead of plastic containers. This was reflected by the slightly lower SI ratios on all three pulse sequences induced by formulation Gl from glass rather than plastic bottles.

The results of this study also showed that for all unacceptable formulations (liquid formulation L2 and granular formulations Gl and G2) , the imaging efficacy was improved by manual shaking. This was reflected by some SI ratios on the second Tl-weighted sequence (i.e., after they had been subjected to an additional manual shaking) being lower than those on the first Tl-weighted sequence (Table 5) . However, improvement by the additional manual shaking was not sufficient to allow these formulations to provide sufficient blackening, since SI ratios were still close to or greater than 2.

Finally, the viscosity of the liquid formulations decreased as a function of the temperature to which they were subjected. This inverse relationship between the viscosity and temperature exposure for the liquid formulations was evident during the one-month exposure of formulation LI, and the one-week exposure of formulations L2 and L3, to various temperatures O 95/20405

(Table 5) . This inverse relationship was not observed with the granular formula ions. The results show that the granular formulations did not always provide good imaging qualities even though their viscosities were greater than 150 cP, confirming that more than high viscosity is needed to produce high quality imaging.

Table 3 G-type Pormtilations Dsed in Compiirative Study

To provide total iron concentration of "150 μg/mL when reconstituted.

Table 4

The ingredients for each formulation are shown in Table 1, and each formulation was imaged once on all three pulse sequences.

Mean*SEM, derived for values measured during the course of study.

Data expressed as mean*SEM, n-2.

Image artifacts (image blurring) .

Ghost artifact observed in one dog.

Table 5 Imaging Results

Storage Signal Intensity Ratio*

Formulations Time Temperature Container Viscosity Proton Tl-

^•c (cP) Tl- T2- Density- Weighted Weighted Weighted Weighted at 20 min

Liquid:

LI 1 month 5 Glass 332 1.0 1.0 1.2 1.23

1 month 30 Glass 297 1.1 .9 1.0 1.43

1 month 40 Glass 254 1.0 .9 .9 1.23

1 month 50 Glass 156 1.1 .8 1.0 1.43

1 week 70 Glass 123 1.1 .8 3 1.3³

L2 1 week 30 Glass 194 1.6 .9 1.6 3.33

1 week 70 Glass 108 5.9 .8 10.1 2.8⁴

L3 1 week 30 Glass 430 1.2 1.0 1.0 1.43

1 week 70 Glass 194 1.2 1.1 1.4 1.3³

Granule:

Gl 6 week 30 Plastic 280 1.9 1.0 2.4 1.7⁴

6 week 50 Plastic 156 2.8 .9 4.2 2.0⁴

6 week 30 Glass 280 1.8 .9 1.7 1.8³

G2 1 week 30 Plastic 236 1.1 .8 1.3 1.6³

1 week 50 Plastic 248 5.2 4.3 16.1 2.7⁴

1 week 70 Plastic 220 3.4 1.4 7.9 3.4⁴

* Signal intensity Ratio of samples to air.

³ Obtained without any additional manual shaking of the samples to evaluate gravitational settling.

4 Obtained after an additional manual shaking of the samples to evaluate the effect of shaking.

Each formulation under the respective circumstances was imaged once on each of the three pulse sequences. No image artifacts were observed.

Table 6 Further Evaluation of the Best Two Formulations

Signal Intensity Ratio

Formulations Storage Viscosity Proton (cP) Density-

Tl -Weighted T2-Weighted Weighted

LI 8 weeks at room temp. 276 1.3 1.2 1.4

1 week at 70°C and 7 weeks at room temp. 102 1.9 1.1 2.3

5 weeks at 70°C and 3 weeks at room temp. 18 12.8* 6.5* 41.8*

L3 8 weeks at room temp. 363 1.1 1.2 1.0

1 week at 70°C and 7 weeks at room temp. 162 1.9 1.0 2.4

5 weeks at 70°C and 3 weeks at room temp. 22 7.4* 1.1 17.0*

Each formulation under the respective circumstances was imaged once on each of the three pulse sequences.

* Image artifacts observed.

-26-

Imaging Performance After Storage

Since formulations LI and L3 were found to provide excellent signal blackening (See the comparative experiments above) they were further evaluated in vitro for performance after storage. The results (Table 6) show that after exposure to room temperature for eight weeks both formulations provided excellent signal blackening (reflected by SI ratios smaller than 2) on all three pulse sequences with no image artifacts. After exposure to 70°C for one week followed by room temperature for seven weeks, both formulations also provided excellent signal blackening (SI ratios below or close to 2) on all three pulse sequences with no image artifacts. However, after exaggerated exposure to 70°C for five weeks plus room temperature for three weeks, the imaging qualities produced by these two formulations were poor (SI ratios greater than 2 and image artifacts observed on at least two of the three pulse sequences) . These results indicate that both formulations behaved similarly to stressful environments in regard to MR imaging.

^■27-

MRI Using the Formulation of the Invention in Humans

Prior to magnetic resonance imaging, human volunteers ingested a contrast agent whose formulation appears in Table 7.

Table 7 Formulation of Contrast Agent

¹ To provide total iron concentration of 150 μg/mL. To adjust viscosity to 310-510 cP. q.s. = sufficient quantity.

Example A

A 25 year old male volunteer orally ingested 500 mL of the formulation described above over a 30 minute time period. Multiple axial abdominal and pelvic Magnetic Resonance (MR) images were obtained prior to ingestion and at multiple time points following ingestion .

The pre-ingestion gradient echo MR images showed intermediate signal intensity within the stomach . On the 60 minute post-ingestion gradient echo MR images , the signal intensity within the stomach had decreased due to the presence of -28-

the formulation (negative contrast effect) and increased distension of the stomach was noted.

Pre-ingestion gradient echo MR images demonstrated moderate to high signal intensity in the lumen of the small bowel which was darkened by the presence of formulation within the lumen of the small bowel on the 60 minute post-ingestion image.

These results show that in this patient 500 mL of product increased signal blackening of both the stomach and small bowel after 60 minutes using a gradient echo pulse sequence. Therefore, the product improved negative contrast and increased visibility of these two organs, when compared to pre-ingestion MR images.

Example B

A 43 year old female volunteer orally ingested 750 mL of the formulation described above over a 60 minute time period. Multiple axial MR images were obtained from the mid-stomach to pelvic region both prior to and following ingestion of the formulation. Images at multiple time periods were acquired following ingestion.

The Tχ-weighted pre-ingestion MR images demonstrated intermediate signal within the stomach, small bowel, and regions of the large bowel. The immediate post-ingestion T__-weighted images demonstrated distension and darkening of the stomach due to the presence of the formulation. In addition, the formulation was seen distributed throughout segments of the small bowel with resultant darkening of the small bowel lumen. The 60 minute post-ingestion Tχ-weighted MR images showed continued darkening of the stomach however, less distension of the stomach was noted. Further distribution and darkening of small bowel segments were noted in the 60 minute post-ingestion images.

In addition, the 60 minute post-ingestion images demonstrated occasional segments of large bowel with intermediate to dark signal intensity.

These results show that 750 mL of product in this patient resulted in an immediate signal blackening of the stomach -29-

using Tl-weighted sequences. Furthermore, 60 minutes after ingestion, excellent negative contrast was seen in the small bowel and occasionally in the large bowel using the same pulse sequence. Therefore, visibility of these areas was greatly improved when compared to pre-ingestion images.

Example C

A 45 year old female volunteer orally ingested 1000 mL of the formulation described above over a 60 minute time period. Multiple axial Ti and T₂~weighted MR images of the abdomen and pelvis were obtained prior to and al various time points following ingestion of the formulation.

The pre-ingestion Tχ-weighted MR images of the lower abdomen and pelvis demonstrated intermediate signal within the lumen of the large bowel and sigmoid colon. The three hour post- ingestion Ti-weighted images demonstrated darkening of the luminal contents of the large bowel, sigmoid colon and rectum.

The T₂~weighted pre-ingestion MR images demonstrated intermediate signal within the stomach, small bowel and large bowel. This intermediate signal was mixed with areas of dark signal within the bowel lumen. The immediate post-ingestion T₂- weighted MR images showed darkening and distension of the stomach. In addition, segments of the small bowel demonstrated areas of darkening compared to the pre-ingestion images representing the presence of the contrast agent formulation.

The three hour post ingestion T₂-weighted MR images demonstrated darkening of segments of large bowel luminal contents compared to the pre-ingestion T₂~weighted MR images. -30-

These results show that overall , 1000 mL ingest ion of product resulted in an increase of signal blackening of luminal contents of large bowel, sigmoid colon and rectum after three hours using Tj- eighted pulse sequences . T₂~weighted pulse sequences, three hours after ingestion, display superior contrast of large bowel luminal contents , as opposed to pre-ingestion images .

In summary, improved negative contrast (MR-imaging efficacy) of various components of the gastrointestinal tract was obtained using three pulse sequences and three doses of the contrast media formulation of the inventions .

The invention has been described in detail with particular reference to certain preferred embodiments thereof, but it will be understood that variations and modifications can be effected within the spirit and scope of the inventions .

Claims

1. A contrast medium comprising a suspension of magnetically responsive particles dispersed in an aqueous carrier comprising one or more substantially hydrated dispersion-enhancing agents.

2. A contrast medium as claimed in claim 1 wherein the dispersion-enhancing agents are xantham gum, microcrystalline cellulose or carboxymethylcellulose sodium.

3. An aqueous composition comprising magnetically responsive particles and one or more substantially hydrated dispersion enhancing agents.

4. A composition as claimed in claim 4 wherein the dispersion-enhancing agents are xantham gum, microcrystalline cellulose or carboxymethylcellulose.

5. A composition as claimed in claim 3 or claim 4, having a viscosity of 25 to 1000 cP.

6. A composition as claimed in claim 5, having a viscosity of 50 to 800 cP.

7. A composition as claimed in any one of claims 3 to

6, wherein the magnetically responsive particles are particles of magnetite, gamma ferric oxide, cobalt, nickel or manganese ferrites.

8. A composition as claimed in any one of claims 3 to

7, wherein the magnetically responsive particles are present in a concentration of from 0.01 to 10 grams per liter.

9. A composition as claimed in claim 8 wherein the magnetically responsive particles are present in a concentration of from 0.1 to 3 grams per liter.

10. A ready-to-use composition comprising magnetically responsive particles and one or more substantially hydrated dispersion enhancing agents.

11. A ready-to-use composition as claimed in claim 10, stored in a container of glass or polyethylene terepthalate.

12. A ready-to-use, stable, low viscosity, fully dispersable composition comprising magnetically responsive particles together with one or more substantially hydrated dispersion enhancing agents.

13. A composition as claimed in claim 12, wherein the magnetically responsive particles are particles of iron oxide.

14. A method of generating a magnetic resonance image of a human or non-human animal body, said method comprising administering to an externally voided body cavity of said body a composition as claimed in any one of claims 3 to 13, and generating a magnetic resonance image of at least a part of said body.

15. Use of a composition as claimed in any one of claims 3 to 13 for the manufacture of an image enhancing contrast medium.