T 1820/19 (Process for preparing a pharmaceutical formulation of contrast agents / GUERBET) 08-11-2022

1. Admittance of D59-D62

The appellant filed D60 and D61 with the grounds of appeal dated 22 August 2019. D62 was filed by respondent 1 with the reply to the grounds of appeal. Lastly, D59, initially submitted during the proceedings before the opposition division, was relied upon by respondent 2 in its reply to the grounds of appeal (see page 18).

The admittance of each of D59-D62 is subject to the provisions of Article 12(4) RPBA 2007.

The Board decided to take into account each of D59-D62 for the following reasons:

- the opposition division did not take any decision as to the admittance of D59. Thus D59 is not a document which could have been presented or was not admitted in the first instance proceedings, such that Article 12(4) RPBA 2007 is not a bar to its admittance.

- In addition, each of D59-D62 is relevant to the questions of unpredictability of the reaction of DOTA with Gd, and of the skilled person's awareness of this fact. The submission of these documents is therefore seen as an appropriate reaction to the reasoning in the appealed decision and as a legitimate attempt by the parties to further support the arguments they had already presented in the first instance proceedings.

2. Main request (patent as granted), inventive step

2.1 The claimed invention

2.1.1 The patent pertains to a process for preparing a liquid pharmaceutical formulation containing gadoteric acid (DOTA-Gd), i.e. the complex formed between the macrocyclic chelate DOTA and the lanthanide gadolinium (Gd). The formulation is used as a contrast agent, especially for magnetic resonance imaging (see paragraph [0001] of the patent).

As explained in the patent (see paragraphs [0003], [0005] and [0010]), in the body, the complexes of chelates with lanthanide are in a situation of chemical equilibrium, which may lead to a risk of undesired release of the toxic free lanthanide. The toxicity of the free chelate (i.e. DOTA which is not under the form of an excipient X[X',L] as defined in claim 1) was also known. In order to solve the problem of tolerance in the patient, the formulation thus contains a very low excess of free chelate DOTA of 0.002-0.4%, so as to increase the scavenging capacity for potential free lanthanide Gd.

In view of this low amount of free DOTA excess, the patent seeks to address the problems of precise and delicate industrial-scale control of the concentrations of free DOTA, and of providing a preparation process ensuring the reliability, reproducibility and stability of the composition, considering in particular the uncertainty of weighing at the industrial scale and the variability of the hygroscopic characteristics of DOTA (see paragraphs [0011]-[0013]).

2.1.2 Claim 1 of the main request accordingly relates to a process for preparing a liquid pharmaceutical formulation containing DOTA-Gd and 0.002-0.4% free DOTA, comprising in summary the following steps:

b) preparation of a liquid pharmaceutical composition by mixing a solution of free DOTA and of free Gd in amounts such that

- all Gd is complexed and

- Cch l > Ct ch l ,

c) measurement in the pharmaceutical formulation obtained in step b) of Cch l,

Clan l being equal to 0;

d) adjustment of Cch l and of Clan l so as to obtain Cch l = Ct ch l and Clan l = 0,

wherein

Cch l is the concentration of free DOTA,

Clan l is the concentration of free Gd,

Ct ch l is the target concentration of free DOTA and is selected in the range of 0.002-0.4%.

2.2 Interpretation of claim 1

2.2.1 According to the appellant, claim 1 allows for the complexation reaction to be incomplete and for free Gd to be present at the end of step b). The amounts recited in step (b) would refer to the input amounts of free DOTA and free Gd entering step b), and would not define a result to be achieved with respect to amounts of free DOTA and free Gd at the output of said step. Furthermore, the subsequent quantitative measurement step c) would be carried out under conditions mimicking what occurred in the reactor upon completion of the reaction, namely by modifying the pH. The contradictory statements in claim 1 would make it necessary to use the description to interpret them, thus leading to the above claim construction, following the principles summarised in the Case Law of the Boards of Appeal (10th edition, 2022, see II.A.6.3.2).

2.2.2 The Board does not adopt the appellant's interpretation, and construes claim 1 as requiring the absence of free Gd in the pharmaceutical formulation obtained in step b).

Firstly, step b) of claim 1 is a step of mixing a solution of free DOTA and of free Gd "so as to obtain complexation of the lanthanide by the macrocyclic chelate". It is therefore not a mere dissolution step, as the appellant argued, but a step in which complexation occurs. Furthermore, step b) mandates that the amounts of free DOTA and free Gd added as starting material are "such that all the lanthanide is complexed". It is therefore a feature of step b) that all Gd is complexed.

Secondly, in the subsequent measurement step c), claim 1 specifies "the concentration of free lanthanide Clan l being equal to 0". Since the wording of step c) clearly relates to a "measurement in the pharmaceutical formulation obtained in step b)", this requirement that the concentration Clan l of free Gd be zero can only apply to the pharmaceutical formulation obtained in step b). The appellant's interpretation, whereby Clan l is different from zero in the formulation obtained in step b) and only becomes zero after modifying the pH of the sample taken from this formulation, is not compatible with the wording of claim 1.

2.2.3 It is a fact that claim 1 still generally defines step b) as a step of preparing a composition containing DOTA-Gd, free DOTA, "and/or free lanthanide". This appears to allow, in one alternative, for the presence of free Gd in the composition resulting from step b), which contradicts the absence of free Gd required by the statements of claim 1 mentioned above (see 2.2.2). However, in the Board's opinion, the skilled reader would resolve this inconsistency by considering that this optional presence of free Gd in the formulation of step b) is ruled out by the subsequent mandatory features of claim 1 mentioned above.

Thus, what appears at first sight to be an inconsistency in claim 1 can be resolved by the skilled reader on the basis of the claim alone. The appellant cannot rely on this inconsistency to give a interpretation to claim 1 which runs counter to its clear technical meaning, namely that the pharmaceutical formulation obtained in step b) contains no free Gd.

2.2.4 In addition, and contrary to the appellant's opinion, this interpretation is not contradicted by the description. The reference example 2 is not according to claim 1, as it is carried out in the presence of an excess Gd in step b), and is thus not relevant to its interpretation. Paragraph [0037] mentions a difference between the amounts of free DOTA and of free Gd added and the stoichiometric proportions, but does not call into question the requirement of claim 1 that this difference be such that all Gd is complexed. Paragraphs [0044]-[0048] pertain to the possibility, in step d), of adjusting the pH so as to shift the reaction equilibrium, but do not indicate that step b) may or must be incomplete or lead to free Gd in the formulation. The description of possible tests for the presence of free DOTA or free Gd, involving preferably a pH 5 buffer medium (see the end of paragraph [0021] and paragraph [0022] on page 7) cannot lead to this conclusion either, because there is no indication in these passages that the testing conditions necessarily modify the tested concentrations or bring the complexation to completion. On the contrary, in the reaction scheme in paragraph [0019], the first step is shown to lead to DOTA-Gd and free DOTA but not to free Gd. Thus paragraph [0019], as well as paragraph [0043], support the Board's interpretation. The appellant's arguments are thus not persuasive.

2.3 The closest prior art and differentiating features

2.3.1 The closest prior art is the information leaflet D1 of the commercial product DOTAREM® 0.5 mmol/ml, solution for injection. This product is indicated for use in magnetic resonance imaging, and contains gadoteric acid, i.e. DOTA-Gd, corresponding to 20.246 g DOTA and 9.062 g gadolinium oxide (i.e. Gd2O3). These amounts correspond to an excess of free DOTA of about 0.12% mol/mol (see paragraph 72 of the appealed decision). The liquid formulation for injection of D1 further contains meglumine and has a pH of 6.5 to 8.0.

Thus, D1 discloses a product obtainable by the process of claim 1 of the main request.

D1 does not disclose any information as to the preparation of DOTAREM®.

2.3.2 The subject-matter of claim 1 of the main request thus differs from D1 by steps b)-d) of the process.

2.4 Objective technical problem

2.4.1 According to the appellant, the claimed process is safe, reliable and accurate, and is suitable for industrial scale production of a pharmaceutical composition according to D1. The appellant further submits that the process is a one pot procedure eliminating the need for isolating the intermediate solid complex prior to formulating the final product, and that it is cost- and time-effective.

However, technical advantages cannot be taken into account in the formulation of the technical problem if it is not made credible that they are indeed achieved as a result of the steps of the claimed process. The fact that the closest prior art D1 does not disclose any preparation process, making a side-by-side comparison with the claimed process impossible, does not mean that any quality alleged by the appellant has to be accepted even in the absence of any evidence. The burden of proof for this rests with the appellant.

The Board finds that no such advantageous technical effect has been shown to arise from the claimed process.

No such evidence is present in the patent or the application as filed. The application as filed (see page 16) equally considered carrying out the complexation step b) with:

- an excess of lanthanide relative to the macrocyclic chelate (i.e. case A), or

- an excess of macrocyclic chelate (case B) leading either to Cch l < Ct ch l or to Cch l > Ct ch l.

The patent however does not comprise any example of the alternative chosen in claim 1 of the main request, where Cch l > Ct ch l. There is thus no basis for the appellant's assertion that starting from a deliberate initial excess of free DOTA above the target is advantageous. The appellant did not explain either why such conclusion should be drawn by analogy from reference example 2.

D37 describes a single reproduction of the process of claim 1 on pilot scale, and is thus not suitable to demonstrate any level of reproducibility or reliability: considering the appellant's argument that the complexation of Gd with DOTA is characterised by a high variability, the single experiment of D37 cannot be regarded as proof that this problem has been solved. Furthmore, D37 is not carried out on industrial scale, and thus does not demonstrate that the claimed process would overcome any difficulty associated with such production quantities.

The protocol followed in D51 does not correspond to the subject-matter defined by claim 1 of the main request. In particular, there is no indication that step b) of D51 led to a formulation in which all Gd is complexed and Cch l > Ct ch l.

D32 does not show any reproduction of the claimed process including the measurement and adjustment steps c) and d).

D56 does not indicate any difference in the outcome of the process when starting from an excess of DOTA or an equimolar amount (see reactions 1, 2 and 3).

No effect as to cost- and time-effectiveness can be acknowledged either: contrary to the appellant's position, carrying out the adjustment step d) by incremental additions is not ruled out by the wording of claim 1.

Lastly, no effect associated with the composition of the formulation itself, such as the narrow range of 0.002-0.4% free DOTA, may be taken into account, because this composition is already known from D1. In particular, no link is shown between any long term storage stability and the steps of the process.

2.4.2 Accordingly, the objective technical problem starting from D1 is to provide a process for making the composition of D1.

2.5 Obviousness

2.5.1 The synthesis of DOTA-Gd by complexation of Gd with DOTA, the toxicity of free Gd and the need to ensure, as in D1, a small excess of ligand in the formulation, were part of the skilled person's common general knowledge (see D8, scheme 3, page 21 lines 7-9 and page 22, chapter 4.2). The textbook D8 further discloses an exemplary one-pot synthesis of a gadolinium complex (see the paragraph spanning pages 20-21, using a related chelate). The skilled person was therefore aware of the possibility of a one-pot process for providing a pharmaceutical composition of a gadolinium-based contrast agent.

The Board considers that the skilled person was also aware that the reaction of DOTA with Gd was unpredictable and that the initially weighted amounts of macrocyclic chelate and Gd did not necessarily result in the intended and calculated ratio, at least because the skilled person knew of the hygroscopic nature of DOTA and the ensuing potential for weighing error (see D58 and D59). In addition, the skilled person would have found confirmation of this fact, in the case of structurally related chelates, in D2 (see page 2424, left column, "Gd-5a Complex (6a)"), D7 (see example 2) and D30 (see example 2).

Thus, the skilled person knew of the critical importance of controlling the concentrations of free DOTA and free Gd in the complex formulation, and the particular uncertainty associated with the complexation reaction, especially in view of DOTA hygroscopicity and weighing error.

The appellant, relying on the post-published evidence D32, D56 and D61, submits that, even taking into account the hygroscopic character of DOTA, the reaction remains unpredictable. The Board considers this not to be decisive. The skilled person knew of the unpredictability of the reaction already for the reasons given above. Therefore, irrespective of further unelucidated factors contributing to this unpredictability, the skilled person would have been prompted to take the required steps when preparing this type of complexes.

2.5.2 Furthermore, it is generally part of the commonly known good manufacturing practices in the pharmaceutical field to control all critical attributes of the manufacturing process (see D13, page 10, section c; see also D57, page 16, point e), in particular to:

- identify and measure critical material and process attributes,

- design a process measurement system to allow real time monitoring of all critical attributes, and

- design process controls that provide adjustments to ensure control of all critical attributes.

2.5.3 The skilled person, starting from the closest prior art D1 and seeking to prepare the DOTA-Gd solution for injection with an excess of free DOTA as indicated therein (see 2.3.1 above), would logically consider carrying out the complexation reaction between Gd and DOTA using an excess of free DOTA reagent. Then, in view of the known critical importance of controlling the concentration of free DOTA in the formulation, and the expected uncertainty of the complexation reaction, the skilled person is led, by its common general knowledge, to carry out steps of measuring (c) and adjusting (d) this concentration.

2.5.4 The sole remaining question is whether the skilled person would use, in the complexation step b), amounts of reagents (free Gd and free DOTA) leading to a concentration of free DOTA Cch l which is higher than the target concentration Ct ch l (the target concentration being here the concentration of free DOTA in the formulation of D1).

The Board notes firstly that using a (deliberately) higher excess of DOTA than necessary in the complexation step b) is not associated with any technical effect (see 2.4.1 above). Furthermore, it follows from the reasoning above that the skilled person would in any case carry out the steps c) and d) of measurement and adjustment, such that the extent of the intermediate excess in free DOTA Cch l would anyway be corrected to the target excess.

Hence the skilled person, which is already led to use an excess of DOTA by the closest prior art D1, would immediately realise that using an excess above the target concentration of D1 would still lead to the same outcome. The Board concludes that the claimed solution is obvious in light of the above common general knowledge.

In these circumstances, a pointer to the specific alternative defined in claim 1 of the main request, whereby Cch l > Ct ch l, is not needed, because this alternative is not associated with any advantageous properties and is equally suited as the others to obtain the desired DOTA-Gd formulation of D1.

Nonetheless, the Board adds that this strategy is illustrated by D2, which describes the preparation of a closely related Gd-macrocyclic chelate complex comprising the steps of mixing an excess of chelate 5a with free Gd (Gd2O3) to obtain complexation, measuring the concentration of free chelate, and adjusting by adding further free Gd to the reaction mixture without resulting in an excess of free Gd. The Board agrees with the appellant that D2 relates to the preparation of a different Gd complex on a laboratory scale. Nonetheless, this document shows that the general approach followed by the present inventors was known in the art.

Accordingly, the subject-matter of the main request does not meet the requirements of Article 56 EPC.

3. Auxiliary requests, inventive step

3.1 It is not necessary to assess the admittance of the main request a2 and auxiliary requests 2, 11-13, 1bis-5bis, 8bis, 9bis, 11bis and 13bis under Article 12(4) RPBA 2007, because these requests can in any case not be allowed for lack of inventive step.

3.2 The alternative "and/or by modifying the pH" in step d) of claim 1 is reformulated in the main requests a1 and a2, and is deleted in the main request b (as well as in auxiliary requests 1bis-13bis). These amendments to step d) are immaterial to the issue of inventive step discussed above.

3.3 In auxiliary requests 2, 2bis, 3, 3bis and 4bis, step b) of claim 1 is amended as follows:

- auxiliary requests 2 and 2bis: "the added amounts of free macrocyclic chelate and of free lanthanide being such that[...]";

- auxiliary requests 3 and 3bis: the mol/mol ratio of the added free DOTA : free Gd amounts is between 1.001 and 1.3;

- auxiliary request 4bis: the amounts of free Gd and of free DOTA added are different from the stoichiometric proportions.

These amendments were introduced in order to exclude an interpretation of claim 1 covering the use of (pseudo-)stoichiometric ratios. They do not modify the assessment of inventive step given for the main request, since the reasoning above is not dependant on such an interpretation. As indicated above (see 2.5.3),

an excess of free DOTA is already indicated in the closest prior art, such that it was obvious to depart from stoichiometric amounts and to perform the complexation reaction with an excess of (added) free DOTA. The excess defined by the free DOTA : free Gd ratio of 1.001-1.3 is not associated with any effect and is arbitrary.

3.4 Claim 1 of auxiliary requests 1bis and 5bis-7bis is identical to claim 1 of the main request b and auxiliary requests 2bis-4bis, such that the same considerations apply.

3.5 The features introduced in claim 1 of auxiliary requests 8, 8bis, 9 and 9bis, mandating that the formulation be injectable or that Gd2O3 be used, do not establish any additional difference over the closest state of the art D1 (see 2.3.1 above).

3.6 Regarding auxiliary requests 10 and 10bis, the feature of claim 1 that the measurement step c) be carried out at pH 7 does not modify the interpretation of claim 1, because the mere indication of this measurement condition does not entail that step b) be incomplete (see 2.2.4 above). This limitation to a pH of 7 for the measuring step is not associated either with any effect on the outcome of the process. The Board agrees with the opposition division that this amendment does not overcome the objections of lack of inventive step (see paragraph 113 of the appealed decision).

3.7 Claim 1 of auxiliary requests 11 and 11bis relates to a process "for preparing an industrial amount of a liquid pharmaceutical formulation". As indicated above for the main request (see 2.4.1), it has not been made credible that the claimed process leads to any technical effect beyond the formation of the formulation of D1, whether on industrial scale or otherwise. Furthermore, the finding of lack of inventive step for the main request is not based on documents limited to smaller amounts, such that the limitation in auxiliary requests 11 and 11bis does not establish an inventive step.

3.8 With respect to auxiliary requests 12 and 12bis, as reasoned in the appealed decision (see paragraph 110), the addition of a step of "adjustment of the pH and of the volume with meglumine", leading to a "formulation of the meglumine salt of a DOTA-gadolinium complex", does not overcome the objection of lack of inventive step, seeing that the composition of D1 already comprises this meglumine salt. The skilled person seeking to produce the formulation of D1 would add meglumine to the formulation, and the addition of this basic compound would modify the pH.

3.9 In claim 1 of auxiliary requests 13 and 13bis, the adjustment step d) is carried out "by eliminating the appropriate amount of free macrocyclic chelate". In the Board's view, where the complexation reaction leads to a relative concentration in free DOTA above that of the target formulation of D1, the elimination, by undefined means, of this excess free DOTA is one way to obtain the desired formulation. As concluded by the opposition division (see paragraph 112 and 112.1 of the appealed decision), this amendment does not modify the assessment given above for the main request.

3.10 Accordingly, none of the auxiliary requests meet the requirements of inventive step.