WO2013186546A1

WO2013186546A1 - Complexes of amphotericin b with poly (glutamic acid)

Info

Publication number: WO2013186546A1
Application number: PCT/GB2013/051525
Authority: WO
Inventors: Karolina LES; Teresa BARATA; Antony Godwin
Original assignee: Polytherics Limited; Drugs For Neglected Diseases Initiative
Priority date: 2012-06-12
Filing date: 2013-06-11
Publication date: 2013-12-19
Also published as: GB201210358D0

Abstract

Complexes The invention provides aprocess for the preparation of a complex of Amphotericin B with poly(glutamic acid), which comprises the steps of i) mixing Amphotericin B with poly(glutamic acid) in the presence of water to form an aqueous solution of a complex of Amphotericin B with poly(glutamic acid); and ii) heating at least a portion of said solution at a temperature in the range of from 45 to 95°C for a period of time from 1 to 72 hours. Also provided is a complex of Amphotericin B withpoly(glutamic acid), which is characterised by (a) a ratio of the peak heights of the UV absorbances of the complex in the regions of 315 and 406 nm of the complex of least 4.5, and (b) a shift of the UV peak of the complex away from 328 nm which is decreased by at least 1 nm compared with the peak shift of the unheated complex.

Description

COMPLEXES OF AMPHOTERICIN B WITH POLY (GLUTAMIC ACID)

Field of the Invention

This invention relates to novel complexes containing Amphotericin B, and a process for making them.

Background of the Invention

Amphotericin B is an antifungal drug, often used intravenously for the treatment of severe systemic fungal infections. Its use, however, presents a number of challenges. First, it is highly insoluble in water, and hence it is difficult to formulate effectively. Secondly, administration can cause severe side-effects, including organ damage, and even death.

The most widely sold formulation of Amphotericin B is the solid deoxycholate salt (Trade Mark Fungizone^®), which immediately prior to use is mixed with water to form a colloidal dispersion for intravenous injection. Another commercially available form is a liposomal preparation, also for intravenous injection, available under the Trade Mark AmBisome^®. Both of these forms have disadvantages. The deoxycholate salt has high toxicity and a narrow therapeutic window. The liposomal formulation has high in vivo activity and relatively low toxicity, but it is very expensive.

It is known that the toxicity of Amphotericin B depends on its state of aggregation, which has been widely studied. For example, Gaboriau et al, Antimicrobial Agents and Chemotherapy, 1997, p. 2345-2351, disclose that heat treatment of the deoxycholate salt Fungizone^® leads to the formation of "superaggregates", and this in turn reduces in vitro toxicity. However, the effect described by Gaboriau et al has proved to be difficult to reproduce, and has not proved capable of practical application. Further, the technique of Gaboriau et al does nothing to solve the problem of the lack of aqueous solubility of the Amphotericin B.

WO 2005/065712 discloses complexes of various pharmaceutical compounds, including Amphotericin B, with a narrow molecular weight distribution polymer that includes units derived from an acrylic acid or a salt thereof. Such complexes assist in the formulation of Amphotericin B by solubilising the compound. However, preparation of such complexes has proved insufficiently robust to be of practical use, and such complexes have relatively low and variable activity in vivo. The polymer is additionally non-biodegradable so is susceptible to accumulation in the body.

We have now found a specific process which results in a novel form of Amphotericin B which has low toxicity, is water soluble, is heat stable, and can be prepared reproducibly and economically.

Summary of the Invention

The present invention provides a process for the preparation of a complex of Amphotericin B with poly(glutamic acid), which comprises the steps of

i) mixing Amphotericin B with poly(glutamic acid) in the presence of water to form an aqueous solution of a complex of Amphotericin B with poly(glutamic acid); and

ii) heating at least a portion of said solution at a temperature in the range of from 45 to 95°C for a period of time from 1 to 72 hours. The complex of Amphotericin B and poly(glutamic acid) resulting from the process of the present invention is novel, and the invention therefore also provides this complex per se. Further, the invention provides pharmaceutical compositions containing the complex together with a pharmaceutically acceptable carrier; a method of treating fungal and protozoal infections which comprises administering a complex or a pharmaceutical composition according to the invention; a complex or a pharmaceutical composition according to the invention for use in therapy, particularly for use in the treatment of fungal and protozoal infections; and a complex according to the invention for use in the manufacture of a medicament for the treatment of fungal and protozoal infections.

Detailed Description of the Invention The process of the invention has been found to produce Amphotericin B in a form which has reduced toxicity. The process involves mixing Amphotericin B with poly(glutamic acid) in the presence of water to form an aqueous solution of a complex of Amphotericin B and poly(glutamic acid), and heating at least part of the solution for a period of time sufficient to reduce the toxicity of the complex. The optimal heating time will depend on the temperature used. Temperatures of at least 45°C have been found to be necessary. The higher the temperature used, the shorter the duration of the heating time needed to obtain optimal results. Preferably step (ii) of the process is carried out at a temperature of at least 50°C. Preferably step (ii) of the process is carried out at a temperature of not more than 95°C, especially not more than 65°C, especially not more than 60°C. For example, step (ii) may be carried out at a temperature of from 45 to 60°C, for example from 50 to 60°C. Preferably the duration of step (ii) is at least 4 hours, and is preferably not more than 48 hours, especially not more than 24 hours. It may for example be from 4 to 24 hours. It is an advantage of the present invention that relatively mild conditions may be used for the heating step;

specifically, relatively low temperatures may be used.

The initial reaction of Amphotericin B with poly(glutamic acid) to form the complex is preferably carried out in the presence of a suitable organic solvent, suitably a water-miscible organic solvent, for example DMSO, DMF, DMA, and a base, preferably an aqueous solution of a base, for example an alkali metal hydroxide, for example NaOH, KOH or CsOH, or a carbonate or bicarbonate, for example NaHC03, Na(C0₃)2 or CS2CO3. In a preferred embodiment, a solution of Amphotericin B and poly(glutamic acid) in an organic solvent is prepared, and an aqueous solution of a base is added, optionally together with additional water. The reaction mixture becomes increasingly aqueous and the Amphotericin B can either associate with the poly(glutamic acid) or precipitate. The base is added to increase the solubility of the poly(glutamic acid) which ensures the poly(glutamic acid) remains in solution. This drives formation of the aqueous-soluble complex of poly(glutamic acid) and Amphotericin B. As the complex is formed, the Amphotericin B remains solubilised as the solution becomes predominantly an aqueous solution. Throughout this specification and claims, an "aqueous solution" should be understood to be a solution in which the solvent is predominantly water, although a minor proportion, for example less than 20%, especially less than 10%), of the solvent, may be a solvent other than water. The reaction mixture may then be heated according to step (ii). In an especially preferred embodiment, this initial aqueous solution of the complex is treated prior to carrying out the heating step (ii) to reduce the quantity of unreacted starting materials, such as free

Amphotericin B, and other impurities, for example organic solvent and base. This may be carried out by any suitable method which can remove free Amphotericin B as well as other impurities. For example, dialysis, gel filtration or ultrafiltration may be used.

Thus, one preferred embodiment of the invention involves mixing Amphotericin B with poly(glutamic acid) in the presence of an organic solvent, water, and a base to form a solution in an aqueous medium of a complex of Amphotericin B and poly(glutamic acid); treating at least a portion of the resulting reaction mixture to reduce the quantity of organic solvent, base and unreacted starting materials; and subsequently heating at least a portion of the resulting aqueous solution at a temperature in the range of from 45 to 95°C for a period of time from 1 to 72 hours.

In general, medicaments containing Amphotericin B are stored in freeze-dried form. It is an important aspect of the invention that heating step (ii) is carried out on the initial aqueous solution of the complex, i.e. on the complex before it is freeze-dried. The mass ratio of Amphotericin B to poly(glutamic acid) in the process according to the invention is typically in the range 0.1 : 1 to 1 : 1. The mixing of Amphotericin B with poly(glutamic acid) and subsequent purification is generally performed at ambient temperature. The content of Amphotericin B relative to poly(glutamic acid) in the complex is not critical, and can vary over a wide range. Typically, it is in the range of from 5 to 60 %wt, for example from 25 to 45 %wt, for example 30 to 40 wt%.

Poly(glutamic acid) exists in two forms. y-Poly(glutamic acid) has the formula

(-CO.CH₂.CH₂.CH(COOH). H-)„, while a-poly(glutamic acid) has the formula

(-CO.CH(CH₂.CH₂.C0₂H). H-)n. Either can be used alone in the process of the invention, or a mixture can be used. The glutamic acid monomer units may be in either R or S optical isomeric form, or a mixture. Poly(glutamic acid) may be linear, branched, hyperbranched or crosslinked. It may for example be in the form of a dendrimer. Any form can be used in the present invention, but preferably a non-crosslinked form is used. Polymer of either narrow MW distribution (Mw/Mn < 1.1) or wide MW distribution (Mw/Mn > 1.1) can be used. The molecular weight of poly(glutamic acid) used is not critical. It may for example be from 30 kDa to 90 kDa, for example from 35 to 60 kDa, for example 45 to 60 kDa.

Complex formation is crucial to the present invention. A complex is an association between two substances in which the bonding is non-covalent. This is distinct from conjugates formed from two substances in which the bonding is covalent. Complexation does not require structural modification of the active agent, in this case Amphotericin B. This is important in the development of new medicinal forms of an active agent that has long been used in humans. The present invention relates to an association between poly(glutamic acid) and Amphotericin B in which the bonding is non-covalent. The complex preparable by the process according to the invention is novel, the

Amphotericin B being present in the form of an aggregate which has not previously been seen in a complex with poly(glutamic acid). Accordingly, the present invention provides a complex preparable by the process of the invention. Specifically, the invention provides a complex of Amphotericin B with poly(glutamic acid), which is characterised by (a) a ratio of the peak heights of the UV absorbance maxima in the regions of 315 and 406 nm of the complex of at least 4.5, and (b) a shift in the wavelength of the maximum of the UV peak of the complex away from 328 nm which is decreased by at least 1 nm compared with the peak shift of the unheated complex. Similarly, the invention also provides a process for the preparation of a complex of Amphotericin B with poly(glutamic acid), which comprises the steps of

ii) heating at least a portion of said solution at a temperature in the range of from 45°C to 95°C for a period of time sufficient to achieve a state of aggregation of the

Amphotericin B in the complex which is characterised by (a) a ratio of the peak heights of the UV absorbance peaks in the regions of 315 and 406 nm of least 4.5: 1, and (b) a shift of the UV absorbance peak maximum away from 328 nm which is decreased by at least 1 nm compared with the peak shift of the unheated complex. It is believed that heating step (ii) changes the aggregation state of the Amphotericin B in the complex. In turn, the aggregation state of the Amphotericin B is believed to correlate with the toxicity of the complex. However, the optimum aggregation state is not simple, and cannot be defined using a single measure. The ratio of the UV absorbance peaks at about 315-328 and 406 nm (A328/A406) is known as the aggregation ratio, and is one indicator of the Amphotericin B aggregation state within the complex. The monomeric form of

Amphotericin B absorbs at about 406 nm and an aggregated form of Amphotericin B absorbs at about 328 nm (see for example Espada et al, Int J Pharm, 361(1-2), 2008, 64-9). The higher the aggregation ratio, the lower the amount of monomeric Amphotericin B present in the complex. However, the type of Amphotericin B aggregation in aqueous solution is also important. This can be measured by the presence of a hyposochromic spectral shift of the peak maximum which would normally occur at about 328 nm. Self-aggregated or dimeric Amphotericin B absorbs strongly at about 328 nm. In higher aggregates, the maximum absorption is shifter to lower wavelengths at <322 nm, for example around 315 nm. The complex formed in step (i) of the present invention will have such a spectral shift, typically to somewhere in the region of 315 nm. It has been found that carrying out step (ii) of the present invention reduces this spectral shift. Neither a particular aggregation ratio nor a particular spectral shift in a complex is sufficient on its own to indicate a complex having reduced toxicity. Rather, the present invention rests on the surprising fact that complexes in which sufficient heating is carried out to ensure that both the degree of aggregation and the type of aggregation, as indicated by the aggregation ratio and the spectral shift respectively, fall within a particular realm, have particularly low toxicity. Once the aqueous solution of the complex has been formed, at least a portion of it is heated, the result being a specific type of aggregate which may be characterised by its UV

absorbance. The complex has an aggregation ratio (conventionally denoted A328/A409, although note that the actual peak will be shifted away from 328 to the region of 315, so the aggregation ratio might be regarded as Α_{3 1}5_-328/Α₄₀9) higher than 4.5, and also has a spectral shift away from 328 nm which is reduced by at least 1 nm compared with the spectral shift away from 328 nm of the complex which has not been subjected to step (ii) of the process of the invention. These two parameters between them define an aggregation state which results in reduced toxicity of the complex. The aggregation ratio is at least 4.5, preferably at least 6, even more preferably at least 8. The spectral shift is reduced by at least 1 nm from the spectral shift of the complex which has not been subjected to a heating step, preferably by at least 1.5 nm, especially by at least 2 nm. Uncomplexed Amphotericin B, exemplified by the commercially available product

Fungizone® once reconstituted in water, exhibits UV absorption peaks with maxima at about 328 and 406. Of course, the precise position of the peaks will depend upon the exact details of the sample and the measurement technique, but in all samples the identity of the peaks is clearly identifiable. When Amphotericin B complex formation with poly(glutamic acid) occurs, the peak at about 328 becomes broadened, and its maximum shifts to lower wavelengths. Nevertheless, the peak is still plainly identifiable as the same peak.

Throughout this specification and claims, it should be understood that all references to peaks at particular wavelengths are references to the maxima of the two peaks, wherever the exact position that those maxima occur.

The complexes of the invention find utility in treating fungal or protozoal infections, for example systemic fungal infections such as aspergillosis and cryptococcosis or protozoan infections such as visceral, mucosal and cutaneous leishmaniasis. The amount of complex which is required to achieve a therapeutic effect will, of course, vary with the route of administration, the subject under treatment, and the particular disorder or disease being treated. It is an advantage of the present invention that the in vivo potency of the

Amphotericin B remains high while the incidence of toxic side effects is much lower. The use of poly(glutamic acid) further ensures that polymer clearance from the body will be achieved, poly(glutamic acid) being biodegradable at physiological pH.

The complex according to the invention is suitably provided as a pharmaceutical composition which comprises a complex according to the invention together with a pharmaceutically acceptable carrier. The complex may be administered by any suitable route, including oral, but since Amphotericin B is known to be an active agent that is not readily absorbed from the gastrointestinal tract, the complex will preferably be administered by a parenteral route, including subcutaneous, intradermal, intramuscular, intravenous [bolus or infusion], and intraarticular. Other non-oral routes of administration may also be used, including inhalation (e.g. using nebulizers or insufflators), rectal, intraperitoneal and topical (including dermal, buccal, sublingual, and intraocular). However parenteral administration is preferred, and a pharmaceutical composition is preferably formulated accordingly.

Formulations for parenteral administration generally include aqueous and non-aqueous sterile solutions which may contain anti-oxidants, buffers, bacteriostats and solutes which render the formulation isotonic with the blood of the intended recipient; and aqueous and non-aqueous sterile suspensions which may include suspending agents and thickening agents. The formulations may be presented in unit-dose or multi-dose containers, for example sealed ampoules and vials, and may be stored in a freeze-dried (lyophilised) condition requiring only the addition of the sterile liquid carrier, for example saline or water-for-injection,

immediately prior to use. Extemporaneous injection solutions and suspensions may be prepared from sterile powders, granules and tablets of the kind previously described.

Exemplary compositions for parenteral administration include injectable solutions or suspensions which can contain, for example, suitable non-toxic, parenterally acceptable diluents or solvents, and optionally other suitable dispersing or wetting and suspending agents.

Whilst a complex of the invention may be used as the sole active ingredient in a

pharmaceutical composition, it is also possible for it to be used in combination with one or more further active agents.

Brief Description of the Drawings

Figure 1 shows the results of testing the products of Example 1, as described in Example 3. Figure 2 shows the results of testing the products of Example 2, as described in Example 3. Figure 3 shows the results of the testing described in Example 3.

Figure 4 shows the UV spectra of (i) Fungizone®; (ii) a complex of Amphotericin B with poly(glutamic acid) prepared without any heating step; and (iii) a complex of Amphotericin B with poly(glutamic acid) prepared with a heating step according to the present invention.

The following Examples illustrate the invention.

Example 1. Preparation of complex of Amphotericin B with poly(glutamic acid). Poly-a-glutamic acid (PG) (100 mg, MW 45 kDa) was allowed to dissolve in dry DMSO (2 mL) overnight with stirring. The polymer solution was then transferred to a 100 mL round bottom flask together with a DMSO solution of Amphotericin B (AmB) (66.6 mg of AmB dissolved in 2 mL of anhydrous DMSO for 1 h). The AmB-PG mixture was left stirring for 1 h at ambient temperature. Under stirring, 1 M aqueous sodium hydroxide (0.773 mL) was added dropwise to the drug mixture followed by a slow addition of 0.19 M NaOH solution (4.1 mL). Additional water (40 mL) was then added immediately following the addition of sodium hydroxide and the resulting mixture was left under continuous stirring for 1 h. The reaction solution was then transferred to a Visking dialysis membrane (MW 12-14 kDa, Medicell International) and dialysis against 1 L of water for 24 h was performed with six changes for fresh 1 L water. After dialysis, the complex solution was divided into aliquots and three aliquots were incubated at either 25 °C, 40 °C or 60 °C for 0 h, 4 h and 24 h. After incubation, the samples were freeze-dried to obtain a solid product.

This product was characterised using UV-Vis spectroscopy as follows: (1) the ratio between absorbance at 328 nm and 409 nm (A328/A409) was calculated; and (2) the difference in the wavelength of maximum absorption away from 328 nm was calculated (at 10 μg/mL of AmB in water). Complexes and standard Amphotericin B were diluted in 50% methanol prior to the analysis. The results are shown in Table 1.

Table 1. AmB aggregation state analysis for AmB-PG complexes prepared in Example

Peak

Temperature of Length of 328 A409

shift

incubation incubation step/h

/nm

na 0 3.9 11.6

4 4.1 11.4

25 °C

24 4.5 11.4 4 4.4 11.2

40 °C

24 4.9 11.0

4 4.6 10.0

60 °C

24 6.3 9.4

Table 1 shows that heating at any temperature up to 60°C increased the aggregation ratio to greater than 4. The shift of the UV peak away from away from 328 nm in the unheated complex was 11.4 nm. Heating at 25 or 40°C for up to 24 h reduced this shift by less than 1 nm, but heating at 60°C for 4 h or more reduced this shift by more than 1 nm.

The haemolytic properties of the products prepared above were evaluated using human erythrocytes isolated from buffy coat residues by density gradient centrifugation using Ficol- Paque™ PLUS. A 5% v/v solution of human erythrocytes was prepared in RPMI 1640. A stock solution of each sample was prepared in sterile water. Non-ionic detergent,

Triton X-100 (Trade Mark) (0.1%) was used as a positive reference for 100% cell lysis. The assay was carried out using concentrations of AmB ranging from 0 to 100 μg/mL. An equal volume of the sample and erythrocytes were aliquot into a 96 well plate and incubated for 24 h at 37°C. Plates were then centrifuged (1500 g x 5 min), supernatants were transferred into new 96-well plates and absorbance was determined at 570 nm by microplate reader (Opsys MR; Dynex Technologies). The degree of lysis was expressed as a percentage of the complete lysis caused by Triton X-100. The results are shown in Figure 1.

Based on knowledge of the heating of the commercial product Fungizone®, it would be expected that increasing the aggregation ratio might reduce toxicity. However, consideration of Figure 1 together with Table 1 shows that consideration of aggregation state measured by the aggregation ratio does not act as a predictor of toxicity. Rather, it is found that a reduction in peak shift is also required. Thus, Figure 1 shows that the haemolytic toxicity of an AmB-PG complex was reduced by heating at 60°C for either 4 h or 24 h. A reduction in haemolytic toxicity did not occur after heating at 25°C or at 40°C.

Example 2. Preparation of complex of Amphotericin B with poly-a-(glutamic acid). Two further batches of the AmB-a-PG complex were prepared as described in Example 1 using two different batches of AmB (labelled batch 1 and batch 2 in Figure 2). Separate aliquots of each complex batch after dialysis were heated at 60 °C for either 0 h, 24 h and 48 h. After heating, the samples were filtered (0.2 μπι) and freeze-dried to obtain a solid product. The aggregation state of the AmB-PG complexes were analysed as in Example 1, and the results are shown in Table 2.

Table 2. AmB aggregation state analysis for PG-AmB complexes prepared in Example 2.

Heating time at 60 °C /h A328 A409 Peak shift /nm

0 3.3 11.6

Batch 1 24 6.7 9.6

48 6.8 9.6

0 2.9 11.8

Batch 2 24 4.8 10.0

48 5.5 9.8

The same biological testing was performed on each sample prepared above as in Example 1 and the results are shown in Figure 2. From Figure 2, it is clear that the toxicity of AmB-PG complex was significantly reduced by introducing a step of heating at 60°C for 24 h or 48 h into the preparative process. The reduction in haemolysis was observed for complexes prepared using both batches of AmB.

Example 3.

AmB-a-PG complexes were prepared and purified by dialysis as previously described in Example 1. After 24 h dialysis, the complex solution was divided into two aliquots. The first aliquot (complex 1) was filtered through a 0.2 μπι sterile filter and freeze-dried to yield a solid product (78 mg). The remaining aliquot (complex 2) was purged with argon and incubated in a water bath at 60°C for 24 h in the dark. After incubation, the sample was filtered using a 0.2 μπι sterile filter and freeze-dried to obtain 73 mg of solid product. The amount of AmB complexed to PG and form of AmB present in the complex was characterised by UV spectroscopy as previously described in Example 1. The results are given in Table 3. Complex 2, prepared with the heating step at 60°C possessed a higher A328/A409 nm ratio than Complex 1, which was not exposed to heating. Figure 4 shows the UV spectra of (i) Fungizone®; (ii) Complex 1; and (iii) Complex 2.

Table 3. AmB aggregation state analysis for AmB-PG complexes prepared in Example 3.

Heating time

Complex ID A328 A409 Peak shift/nm

at 60°C /h

Complex 1

(comparative) 0 3.2 11.6

Complex 2 24 5.6 9.2 The haemolytic properties of Complexes 1 and 2 were determined as previously described in Example 1 and the results shown in Figure 3. The haemolytic toxicity of AmB-PG complex was reduced from 35% (Complex 1) to 2% (Complex 2) by heating at 60°C for 24 h.

The in vivo activity of Complex 2 against Leishmania donovani HU3 in BALB/c mice was tested. The complex was dosed intravenously on day 7, 9 and 11 post infection at 1 mg/kg per injection based on AmB loading. On day 14, mice were sacrificed and the level of infection was established by microscopic examination of liver sections stained with Giemsa. An in vivo inhibition of 91% (±6) was obtained. When the experiment was repeated using a dose of 10 mg/kg per injection, an in vivo inhibition of 100%) was obtained. When the experiment was repeated using Complex 1 (i.e. complex prepared without a heating step), severe toxicity in mice was observed following a single dose of 9 mg/kg. Such toxicity had not been observed with Complex 2.

Example 4: Preparation of complex of Amphotericin B with poly-y-glutamic acid. ΑιηΒ-γ-PG complex was prepared and purified by dialysis as described in Example 1, with the exception that poly-y-glutamic acid was used instead of poly-a-glutamic acid. After dialysis, the complex solution was argon purged and incubated at 60 °C for 24 h. After incubation, the sample was filtered through a 0.2 μπι sterile filter and freeze-dried to yield a solid product (155 mg). The aggregation state of the AmB- γ-PG complex and its in vitro toxicity to human RBCs were analysed as in Example 1. The results are shown in Table 4.

Table 4. Aggregation state and haemolytic properties of AmB-y-PG complex.

% haemolysis at 100 AmB content A₃₂₈/A409 _{Peak μ§/ηΛ after 24 h} j-^o^j shift /nm incubation with RBCs

[%]

AmB-Y-PG

complex

The complex prepared using γ-PG possessed similar physicochemical properties to complexes prepared from a-PG, and was not haemolytic to human RBCs.

Claims

1. A process for the preparation of a complex of Amphotericin B with poly(glutamic acid), which comprises the steps of

ii) heating at least a portion of said solution at a temperature in the range of from 45 to 95°C for a period of time from 1 to 72 hours.

2. A process as claimed in claim 1, in which step (ii) of the process is carried out at a temperature of at least 50°C.

3. A process as claimed in either claim 1 or claim 2, in which step (ii) of the process is carried out at a temperature of not more than 60°C.

4. A process as claimed in any one of the preceding claims, in which the duration of step (ii) is at least 4 hours.

5. A process as claimed in any one of the preceding claims, in which the duration of step (ii) is not more than 24 hours.

6. A process as claimed in claim 1, in which step (ii) of the process is carried out at a temperature of from 50 to 60°C for from 4 to 24 hours.

7. A process as claimed in any one of the preceding claims, in which step (i) is carried out in the presence of an organic solvent and an aqueous base.

8. A process as claimed in claim 7, in which a solution of Amphotericin B and poly(glutamic acid) in an organic solvent is prepared, and an aqueous solution of a base is added, optionally together with additional water.

9. A process as claimed in any one of the preceding claims, which comprises mixing Amphotericin B with poly(glutamic acid) in the presence of an organic solvent, water, and a base to form a solution in an aqueous medium of a complex of Amphotericin B and poly(glutamic acid); treating at least a portion of the resulting reaction mixture to reduce the quantity of organic solvent, base and unreacted starting materials; and subsequently heating at least a portion of the resulting aqueous solution.

10. A process for the preparation of a complex of Amphotericin B with poly(glutamic acid), which comprises the steps of

ii) heating at least a portion of said solution at a temperature in the range of from

45°C to 95°C for a period of time sufficient to achieve a state of aggregation of the

Amphotericin B in the complex which is characterised by (a) a ratio of the peak heights of the UV absorbances in the regions of 315 and 406 nm of least 4.5: 1, and (b) a shift of the UV absorbance peak maximum away from 328 nm which is decreased by at least 1 nm compared with the peak shift of the unheated complex.

11. A process as claimed in claim 10, carried out as defined in any one of claims 1 to 9.

12. A complex of Amphotericin B with poly(glutamic acid), which is characterised by (a) a ratio of the peak heights of the UV absorbances of the complex in the regions of 315 and 406 nm of the complex of least 4.5, and (b) a shift of the UV peak of the complex away from 328 nm which is decreased by at least 1 nm compared with the peak shift of the unheated complex.

13. A complex as claimed in claim 12, in which the ratio of the UV absorbances in the regions of 315 and 406 nm of the complex is at least 6, and the shift of the UV peak away from 328 nm of the complex is at least 1.5 nm.

14. A complex of Amphotericin B and poly(glutamic acid) preparable by a process as claimed in any one of claims 1 to 11.

15. A pharmaceutical composition comprising a complex as claimed in any one of claims 12 to 14, together with a pharmaceutically acceptable carrier.

16. A pharmaceutical composition as claimed in claim 15, suitable for administration by a parenteral route.

17. A pharmaceutical composition as claimed in either claim 15 or claim 16, which also comprises a further active agent.

18. A method of treating a fungal or protozoal infection, which comprises administering a complex as claimed in any one of claims 12 to 14 or a pharmaceutical composition as claimed in any one of claims 15 to 17.

19. A complex as claimed in any one of claims 12 to 14 or a pharmaceutical composition as claimed in any one of claims 15 to 17, for use in therapy.

20. A complex as claimed in any one of claims 12 to 14 or a pharmaceutical composition as claimed in any one of claims 15 to 17, for use in the manufacture of a medicament for the treatment of fungal and protozoal infections.