Another Italian take on the Fulvestrant Saga. The Court of Milan on technical prejudice, plausibility and off-label use.

AstraZeneca has notoriously fought its fulvestrant patent portfolio all across Europe, including Italy, for quite some time (see for instance here, here, here and here). Fulvestrant is an oncological product used for the treatment of breast cancer and is marketed by AstraZenca as Faslodex.

The Court of Milan recently chipped in again with its decision of 3 December 2020 (No. 7930/2020, Judge Rapporteur Ms Alima Zana, AstraZeneca v. Teva, here), delivering a number of interesting points on validity and infringement of second medical use patents.

In particular, the Court argued that the Italian arm of AstraZenca’s European patent No. EP 1 272 195 (“EP195”) lacked inventive step, among other things, as it did not overcome a technical prejudice, that the claimed therapeutic indication was not plausible and that – in any case – AstraZeneca did not offer sufficient evidence of Teva’s off-label infringement.

Photo by Chokniti Khongchum on Pexels.com

Procedural background

In 2018, AstraZeneca filed an action against Teva before the Court of Milan for the alleged infringement of the EP195 patent, titled “use of fulvestrant in the treatment of resistant breast cancer”. Teva counter-claimed arguing that EP195 was invalid.

In May 2020, as Teva was about to launch Fulvestrant Teva into the Italian market, AstraZeneca filed a petition for preliminary injunction based on EP195. The Court refused the PI in August (here) as it deemed that AstraZeneca’s action did not meet the “likelihood of success on the merits” requirement (so-called fumus boni iuris). The Court thus went on to decide the case on the merits.

The patent and the allegedly infringing product

Claim 1 of EP195 reads: “Use of fulvestrant in the preparation of a medicament for the treatment of a patient with breast cancer who previously has been treated with an aromatase inhibitor and tamoxifen and has failed with such previous treatment” – a typical “Swiss-type” claim (for additional information on “Swiss-type” claims, see EPO’s Guidelines, here).

We gather that the patented use of fulvestrant (in “the treatment of a patient with breast cancer who previously has been treated with an aromatase inhibitor and tamoxifen and has failed with such previous treatment”) was not indicated in the SmPC or package leaflet of Fulvestrant Teva. The product’s marketing authorization (“MA”) would confirm this (see the 2016 version of the MA, here, and further updates here, here, here and here).

Expert opinions and foreign decisions

The Court of Milan deemed that EP195 was invalid, going against the opinion of its own Court-appointed expert. This is rather uncommon in Italian patent proceedings since Judges have no technical background and they necessarily rely on their experts’ reports. Here, however, the Court was convinced by the opinion issued by a different expert in the parallel proceedings brought by AstraZeneca against another generic company on EP195 (Docket No. 37181/2018 – this case is referred to in the judgement, but we are not aware of a published decision yet).

Also, the Court relied on the foreign decisions issued on EP195 by the German Bundespatentgericht (here), the Swiss Bundespatentgericht and the Court of Barcelona (order of 18 July 2018, confirmed in 2019, here).

In doing so, the Court stressed that judges can base their decisions also on evidence that is acquired in different proceedings between the same parties, or even different parties. The general rule of Article 116 of the Civil Procedure Code (CPC) – according to which “the judge must assess the evidence according to his or her prudent judgment, unless the law provides otherwise” – fully applies to the Expert Report issued in the parallel proceedings brought by AstraZeneca on EP195. Besides, the “parallel” Expert Report was also filed by the parties to these proceedings and heavily discussed in their briefs, so that there was no surprise argument.

Lack of inventive step

The Court-appointed Expert deemed that the closest prior art to EP195 was the “Gale” article. The Court picked the “Vogel” article instead as the closest prior art and determined the objective technical problem accordingly. Applying the EPO’s “could-would approach”, the Court of Milan concluded that the solution disclosed in EP195 (i.e., to use fulvestrant as a line of treatment for breast cancer after aromatase inhibitors and tamoxifen) lacked inventive step.

According to the Court, the person skilled in the art would have found in “Vogel” an incentive to use fulvestrant as a third line treatment, after tamoxifen and aromatase inhibitors, since (i) “Vogel” mentioned pure anti-estrogens as emerging candidates to be used in the second-to-fourth therapeutic line (ii) fulvestrant was identified as a pure anti-estrogen in a publication (“DeFriend”) referenced in “Vogel” and (iii) “Vogel” specifically taught that pure anti-estrogens inhibit the growth of a tumor already treated with tamoxifen.

The Court also denied that the use of fulvestrant after tamoxifen overcame a technical prejudice. AstraZeneca argued that because tamoxifen and fulvestrant have a similar mechanism of action, the person skilled in the art would have feared that the breast cancer may have formed resistances and mutations to fulvestrant after administration of tamoxifen. The Court, however, found that tamoxifen and fulvestrant have substantial differences, which were known in the art, and therefore no technical prejudice could arise. On this point, the Court directly referenced the German Bundespatentgericht decision (here), which found that there was no technical prejudice against the choice of fulvestrant as a pure anti-estrogen since:

the person skilled in the art associated the choice of fulvestrant with a reasonable expectation of success because Vogel considers it to be an advantageous example of a pure anti-estrogen, as it does not show any relevant side effects despite its high efficacy and shows fewer side effects than other drugs known to be effective for the endocrine treatment of breast cancer”.

Plausibility (?)

As an additional ground of invalidity, the Court of Milan stressed that the patent would be valid only in relation to a purely palliative treatment (and not as an adjuvant therapy) in a patient affected by breast cancer that was previously treated by an aromatase inhibitor and tamoxifen, where said treatments were unsuccessful. This is because the experimental data mentioned in the patent allegedly “refer only to a palliative use of fulvestrant”. Therefore, these data would show that a “plausible solution to the technical problem was found only in relation to the palliative use” of the active principle. The Court thus held that, in any case, AstraZeneca would have had to submit a limited wording of claim 1.

The Court, however, did not precisely identify which patentability requirement would have been affected by this “plausibility” issue, although novelty, inventive step and sufficiency of disclosure can all be theoretically impacted by a (lack of) plausibility argument (on the “nebulous” plausibility requirement, see a recent article by Rt. Hon. Professor Sir Robin Jacob, here). The Court was arguably hinting at an insufficiency of disclosure in relation to the therapeutic indication. It should be stressed, however, that the EPO does not expect applicants to submit clinical trials for each therapeutic indication. In particular, the EPO Guidelines (here) suggest that:

Either the application must provide suitable evidence for the claimed therapeutic effect or it must be derivable from the prior art or common general knowledge. The disclosure of experimental results in the application is not always required to establish sufficiency, in particular if the application discloses a plausible technical concept and there are no substantiated doubts that the claimed concept can be put into practice (T 950/13 citing T 578/06). Any kind of experimental data have been accepted by the boards. It has also been repeatedly emphasised that ‘it is not always necessary that results of applying the claimed composition in clinical trials, or at least to animals are reported’ (T 1273/09 citing T 609/02)”.

Besides, in an earlier case on fulvestrant (Decision of 24 July 2019, Actavis v. AstraZeneca, concerning the Italian portion of EP 2 266 573: here), the Court of Milan had rejected an insufficiency/plausibility argument by stating that: “The inclusion of examples in the patent specification is not mandatory nor is it a requirement for the sufficiency of the disclosure the completion of a clinical trial at the date of the filing, as it is possible to rely on subsequent trials (decision T 433/05), provided that the effect was plausible at the date of filing” (for a comment on this case see here).

Lack of infringement

To seal the deal, the Court of Milan then went on to say that, in any case, Teva’s product would not have infringed EP195. The Court argument goes as follows:

I. The asserted patent is a patent for a second medical use, the patent being directed at the protection of a known substance, limited to the use claimed;

II. In such a case the infringement cannot be found if the generic medicine in its marketing authorisation contains a clear limitation to uses that do not violate patent rights […]

III. In order for there to be infringement […] it is necessary that the [competitor’s] product is not only abstractly suitable for the claimed use, but that the package leaflet indicates the use of the product for the intended purpose.

This assessment is supported by the numerous proceedings in other jurisdictions that have ruled in favour of non-infringement [e.g. the decision of the Oberlandesgericht Düsseldorf, 9 January 2019, mentioned in the footnote].

The defendant’s argument that Teva’s drug is not ‘suitable’ for the claimed use is therefore correct, as this is not the case from a regulatory (since the product is not authorised for the claimed use) nor from a practical point of view (since the claimed use is not foreseen or included in the oncological practice), nor has the [plaintiff] offered any evidence to the contrary.

Due to the finding of invalidity of EP195, the Court’s non-infringement argument is ad abundantiam and consequently rather short. However, some passages are worth being discussed.

First, as to point (II), it is not entirely clear what the Court is referring to when it says that the MA of Fulvestrant Teva “contains a clear limitation to uses that do not violate patent rights”. Assuming that the Fulvestrant Teva’s MA does not explicitly include the therapeutic indication covered by EP195, the Court could be referring to the following passage in the authorization (see here and here):

The holder of the marketing authorization for the generic drug is solely responsible for full compliance with the industrial property rights of the reference medicinal product and of the patent legislation in force.

The holder of the MA of the equivalent medicine is also responsible for ensuring full compliance with the provisions of Article 14(2) of Legislative Decree 219/2006, which requires that those parts of the summary of product characteristics of the reference medicine which refer to indications or dosages still covered by a patent at the time the medicine is placed on the market are not included in the leaflets”.

If this is the case, the Court’s argument on the MA is not particularly convincing. Indeed, this wording – which is boilerplate and often included in MAs – can hardly be read as a “clear limitation to uses that do not violate patent rights”. Infringing uses are technically neither excluded nor forbidden by it, at least under patent law. The Italian Pharmaceutical Authority simply stresses that it is the generics’ responsibility to steer clear of patent rights.

Second, as to point (III), the Court would seem to suggest that patent infringement of a second medical use patent can only occur if the package leaflet “indicates the [claimed] use“. However, the fact that a specific therapeutic indication is missing from the MA, SmPC or package leaflet does not exclude per se that the drug may be used and prescribed for the patented use.

Photo by Pixabay

Besides, the Court’s argument is not fully supported by the case law of the Oberlandesgericht Düsseldorf referred to in the decision (see here and here). Although in the parallel AstraZeneca v. Teva proceedings the Düsseldorf Court found that EP195 was not infringed on the merits of the case, German case law has established that patent infringement can occur also in “off-label” cases, where the pharmaceutical product is (i) suitable to be used according to the patented indication and (ii) the manufacturer or distributor takes advantage of circumstances that ensure that the product is sold and/or is used for the intended purpose. This, in turn, requires a sufficient extent of use in accordance with the patent as well as knowledge (or at least a willful blindness) of this use on the part of the manufacturer or distributor (see Oberlandesgericht Düsseldorf, 5 May 2017, I-2 W 6/17, Estrogen Blocker, § 85, another case on EP195, here).

In any way, the Court of Milan somewhat softens its argument by adding, in the last paragraph, that AstraZeneca did not provide any evidence of the infringement by Teva or that the use of fulvestrant according to EP195 was the current oncological practice in Italy. Conversely – one may wonder – if AstraZeneca had provided sufficient evidence of the off-label use a finding of infringement could have been possible.

For further answers we now look forward to the decision of the Court of Milan in the parallel case (Docket No. 37181/2018) and will report on any interesting development.

Giovanni Trabucco

Court of Milan, 3 December 2020, Decision No. 7930/2020, AstraZeneca v. Teva

Facebook’s (failed) vaccine against the Infodemic

Alongside the Covid-19 pandemic, an equally dangerous emergency is underway. It relates to the circulation of false information on the internet and particularly on social media platforms. This is a pervasive and worldwide phenomenon which calls into question the role that digital platforms should play in tackling disinformation and misinformation. The question is in fact: should digital platforms be in charge of addressing the problem of online disinformation?

Last year Avaaz conducted a study aimed at monitoring the spreading on Facebook of misleading content related to the pandemic. It was also aimed at analysing and evaluating the effectiveness of the big tech policies to combat this “Infodemic”. The results showed an extreme fallacy and a decisive delay in the implementation of relevant policies. A year later, the organization published a second study that returns to this issue in order to compare those data with the current situation and to verify if an improvement occurred.

In practical terms, intervening to stop an alleged false content from circulating on social media is something left to the self-regulation of the platform. According to the said internal practice, the information must be submitted to the so-called fact checkers, that verify and certify that it is actually false information. In other words, the fake news must be “debunked”. This work is carried out by specialized companies that can be partner of the platform or independent from it. Once the fact check has been notified (before receiving a confirmation, in fact, the social media do not intervene independently) Facebook can take measures consisting of either the labelling or the removal.

The Avaaz study examined a sample of 135 Facebook content in five different languages (English, French, Italian, Spanish and Portuguese) identified as fake by independent fact checkers. A number of key findings emerged from this sample. The first is that the most widespread narrative concerns vaccines side effects, including death. Secondly, it appears that Facebook is more reluctant to intervene with fake content in non-English language. This results either in a late intervention (30 days for non-English content compared to 24 days for English-language false content) or, even, in a lack of intervention with the effect that European citizens would seem to be more exposed to the risk of misinformation than Americans (or, properly, than English-speaking countries citizens).

Ultimately, the study highlights how Facebook’s policies relating to Covid-19 disinformation and misinformation in Europe should be reviewed and strengthened especially in time of global crisis. In this regard, it should be noted that the EU Commission, even before the pandemic, was committed to fight the spread of the phenomenon at stake. In particular, in 2018 a Code of Practice on Disinformation (“the Code”) was adopted and signed by online platforms, the leading social networks and advertisers. The Code aims to implement the 2018 Communication from the Commission and it also identifies some best practices. However, Avaaz suggests that this document should be amended, providing, as an example, the following measures: i) a retroactive notification for users who interacted with fake content; ii) the reduction of the acceleration of harmful content caused by the algorithm and, finally; iii) the establishment of an independent monitoring regulator.

It should also be noted that the new proposed Digital Services Act (“DSA”) aims specifically to tackle the spread of fake news online and to increase transparency providing a series of obligations for online platforms, including social media. It will therefore be interesting to see how the works will evolve. In the meantime, one thing is certain: institutions have not found a vaccine against the Infodemic… so far.

Valeria Caforio

What’s Up, WhatsApp?!

In a GDPR urgency proceeding, the Hamburg Commissioner for Data Protection and Freedom of Information (HmbBfDI), Johannes Caspar, has taken action against Facebook. The aim of this proceeding – whose decision is expected before May 15, 2021 – is to comprehensively protect WhatsApp users in Germany, who are confronted with the company’s new terms of use. Against this backdrop, May 15 can be emphasized as an important deadline, since by that date users must consent to data processing by the parent company Facebook. The fear is, that the data will be used in particular for marketing purposes, which goes beyond the scope of analysis and security.

After Facebook had announced the new terms and conditions at the beginning of this year, a discussion arouse. As a result, the company decided to postpone the introduction to May. With many million Whatsapp users in Germany alone, Johannes Caspar now stressed the importance of having functioning institutions in place, in order to prevent the misuse of data power. At this point, Caspar could not exclude, that the data-sharing provisions between WhatsApp and Facebook would be enforced illegally, due to the lack of voluntary and informed consent. In order to prevent a potentially unlawful exchange of data and to put an end to any impermissible pressure on users for giving their consent, the formal administrative procedure was initiated.

Based on Art. 66 GDPR (“exceptional circumstances”), the emergency procedure is aimed at the European headquarters in Ireland. The American company is given the opportunity to state its position, whereby it can be expected that Facebook will consider the adjustments to be sufficient. However, the Hamburg data protection authorities had already issued an injunction against such data matching in 2016. Although Facebook took legal action, the company did not prevail in court (OVG Hamburg, February 26, 2018 – 5 Bs 93/17 – K&R 2018, 282).

The outcome of the proceedings in Hamburg is eagerly awaited since it may have an impact on the entire European market, given the direct applicability of Art. 66 GDPR in the different Member States. Although the decision from 2018 could delineate a trend, the outcome is open.

Dario Henri Haux

See the media statement at: https://datenschutz-hamburg.de/pressemitteilungen/2021/04/2021-04-13-facebook

Trumping the First Amendment: Updates on the Twitter Saga

The decisions concerning the blocking of Twitter and Facebook accounts belonging to Donald Trump are still pending. 

On the contrary, the judicial proceedings regarding the attempts by then-President Trump himself to limit the reactions to his own posts by other Twitter users was decided by the United States Court of Appeals for the Second Circuit defining the question of whether blocking another social media user could consist in a violation of the First Amendment.

President Trump, acting in his official capacity as President of the United States, petitioned for a writ of certiorari in August 2020. Because, following the elections, President Biden would have become the petitioner to this action, the Justice Department asked the Supreme Court to declare the case moot (see here). On 5 April 2020, the US Supreme Court, following its established practice, granted the writ of certiorari, vacated the lower judgement, and remanded the case to the Court of Appeals for the Second Circuit with instructions to dismiss the case as moot.

What makes this decision particularly interesting is Justice Thomas’ concurring opinion.

Indeed, the concurring opinion of Justice Thomas addresses the ongoing discussions on the nature of “digital spaces”, in between private and public spheres. Many see an inherent vice in a system based on a private enforcement of fundamental rights to be performed by platforms, such as Twitter. 

In the case at stake, the petition revolved around the possibility to qualify a Twitter thread as a public forum, protected by the First Amendment. At the same time, however, the oddity of such a qualification becomes clear, as Twitter – a private company – has unrestricted authority to moderate the threads, according to its own terms of service.

Against this backdrop, Justice Thomas states that: “We will soon have no choice but to address how our legal doctrines apply to highly concentrated, privately owned information infrastructure such as digital platforms”.

From what has been said here, it should be clear that a turning point in the ongoing discussion that heats scholars, legislators as well as all players on a global scale is marked. Indeed, Justice Thomas analyses the legal qualification of the “public forum” doctrine and its applicability to Twitter threads since the main controversial element of the case regards the existence of a governmental control of the digital space (even in the limit of a single Twitter account). 

Given that Mr. Trump often used his personal account to speak in his official capacity, it is questionable whether this element might be fulfilled in the case at stake. At the same time, the private nature of providers with control over online content, combined with the concentration of platforms limiting the number of services available to the public, may offer new ways of legally addressing these challenges. For example, Justice Thomas proposes to consider the doctrines pertaining to limitations to the right of a private company to exclude others, such as “common carriers” or “public accommodation”. 

In this regard, Justice Thomas found that: “there is a fair argument that some digital platforms are sufficiently akin to common carriers or places of accommodation to be regulated in this manner”, especially in cases where digital platforms have dominant market share deriving from their network size.

Interestingly, Justice Thomas did not miss the chance to depict the digital environment as such, when stating that: “The Internet, of course, is a network. But these digital platforms are networks within that network”.

What is more, the dominant position of the main platforms in the digital market is taken for granted without further analysis. Namely, it is stressed that the existing concentration gives few private players “enormous control over speech.” This is particularly valid, considering that viable alternatives to the services offered by GAFAM are barely existing, also given to strategic acquisitions of promising start-ups and competitors. 

It becomes evident that public control over the platform’s right to exclude should – at least – be considered. In that case, platforms’ unilateral control would be reduced to the benefit of an increasing public oversight, that implies: “a government official’s account begins to better resemble a ‘government-controlled space’.”

Justice Thomas highlights that this precise reasoning gives strong arguments to support a regulation of digital platforms that addresses public concerns.

In conclusion, according to Justice Thomas, the tension between ownership and the right to exclude in respect to the right of free speech must be solved expeditiously and consideration must be given to both the risks associated with a public authority (as then-President Trump) cutting off citizens’ free speech using Twitter features, and the smoothing power of dominant digital platforms. 

Andrea Giulia Monteleone

The full text of the Supreme Court decision: 20-197 Biden v. Knight First Amendment Institute at Columbia Univ. (04/05/2021) (supremecourt.gov)

A Puzzling Decision: Freedom on Toy Blocks’ Production Market vs. LEGO’s Protection

We already referred to a case concerning Lego, and, in particular, concerning Lego mini-figures and their possibility to be registered as shape marks (see here).

Now, after filing an application for registration of a Community Design with the European Union Intellectual Property Office (EUIPO) to be applied in Class 21.01 of the Locarno Agreement of 8 October 1968 with the following description: “Building blocks from a toy building set”, Lego A/S (“Lego”) got a big surprise by Delta Sport Handelskontor GmbH (“Delta”) which decided to “raise its concerns”.

Specifically, on December 8, 2016, Delta decided to challenge the validity of Lego’s application on the basis of the fact that all the features of appearance of the product concerned by the contested design were solely dictated by the technical function of the product and, for that reason, were excluded from protection; the claims were mostly grounded on Article 25(1)(b) of Regulation No 6/2002, along with Articles 4 to 9 of that same Regulation.

In the first instance, on October 30, 2017, EUIPO’s Cancellation Division rejected the invalidity claim while, in appeal, the ruling was overturned and the design was invalidated (by decision of the Board of Appeal – April 10, 2019) (texts of both decisions are available here under section “Decisions”).

At this point, Lego decided to bring the case in front of EU’s General Court claiming (i) the annulment of the contested decision and (ii) the upholding of the decision of the Cancellation Division.

The EU’s General Court, on March 24, 2021, annulled the appeal decision, specifying, inter alia, what follows (full text available here):

  • The Board of Appeal failed to assess whether the design met the requirements of the exception provided for in Article 8(3) of Regulation No 6/2002 for which “the mechanical fittings of modular products may nevertheless constitute an important element of the innovative characteristics of modular products and present a major marketing asset, and therefore should be eligible for protection”. Since it failed to do so, it erred in law.
  • In order to assess whether a design product should be invalidated, all the features and/or elements of the same should be dictated by technical function, but if at least one of the features is not imposed exclusively by technical functions, the design cannot be declared invalid. In that regard, the General Court notes that the fact that the LEGO brick has a smooth surface of the upper face of the product is not present among the characteristics identified by the Board of Appeal, even though it is a feature of the appearance of the brick itself; and that
  • with reference to the above, the burden of proof should rely on the applicant of the invalidation query and then be ascertained by EUIPO.

Maria Di Gravio and Francesca Di Lazzaro

General Court (Second Chamber), 24 March 2021, LEGO A/S vs EUIPO, Case T‑515/19

HAPPY EASTER EVERYONE FROM IPlens

The Italian Competition Authority clarifies how to improve the internal market competitiveness. Focus on digital platforms.

On 23 March, the Italian Competition Authority (ICA) published a 105 pages report providing the Italian Government with a number of pro-competitive reform proposals in view of the forthcoming annual Competition Act (full report in Italian here). This last can be read as the outcome of a constructive dialogue between the ICA and the Government, whereby the first clarifies how to unlock the Italian market competitiveness and the latter may turn these suggestions into binding law. It falls outside the scope of this summary to examine whether, since the entry into force of the Competition Act in 2009, the Government year by year in charge followed the advices of the ICA. In any event, the recent call of Prime Minister Draghi himself for the ICA opinion during the opening speech at the Italian Parliament, bodes well for an effective 2021 Competition Act.

Covering a wide range of topics, the report overall implies the ICA embraced the new EU wide zeitgeist of the coming years, namely, sustainability and digitalization.

The reform proposals focusing on the connection between competition law and sustainable development, indeed, echo and tailor to the Italian markets’ peculiarities the innovative studies released by the Dutch and Greek Competition Authorities (and the recent report of the CMA). As for the digital sector, two proposals for legislative amendments concerning, first of all, the economic dependence law and, secondly, an ex-ante regulatory framework for digital platforms, stand out for the radical impact on the ICA enforcement regime they could have.

With regard to the first point, the Italian Law 192/1998 (“Law on Subcontracting in Production Activities”) imposes companies with a strong bargaining power against customers and/or other companies to avoid any abuse of this situation and related economic dependence on the latter. The economic dependence is determined according to the following criteria:

  1. the possibility for the company to impose an “excessive imbalance of rights and obligations in the companies’ commercial relationships”, and
  2. the “effective possibility”for the alleged abused company in“finding satisfactory alternatives on the market.”

The available remedies against this abuse consist in declaration of nullity of the agreement concerned, injunctions and compensation for damages. Remarkably, within this assessment, it finds no room neither the (costly and time consuming) antitrust analysis of firms’ dominance in a given market nor the assessment of anticompetitive effects herein occurred. Conversely, it shares with the abuse of dominance legal institute the up to 10% fines of the turnover of the company concerned the ICA may impose and the penalty payments in case of non-compliance with the ICA’s decision.

Against this background, an emerging trend pioneered by new legislation adopted in Germany and Belgium and by a more aggressive enforcement by the France antitrust watchdog, is revitalizing the legal instrument at hand, sharpening its teeth.

The ICA is following this trend setting its sight on digital platforms. The core of the proposal rests upon the introduction of a rebuttable presumption of economic dependence  “in case a company make use of the intermediary services provided by a digital platform holding (…), a crucial role in reaching end users and suppliers” (also due to network effects and/or data availability it can leverage ). A list of non-exclusive forbidden conducts supplements this proposal, fattening this newly shaped notion of abuse of economic dependence and enhancing the legal certainty for digital platforms. These are:

  1. directly or indirectly imposing unfair purchase or selling prices, other unfair conditions or retroactive non-contractual obligations;
  2. applying dissimilar conditions to equivalent transactions;
  3. refusing products or services’ interoperability or data portability, thereby harming competition;
  4. subordinating the conclusion and execution of contracts and/or the continuity and regularity of the resultingcommercial relations to the acceptance by other parties of supplementary obligations which, by their very nature and according to commercial usage, have no connection with the subject of such contracts;
  5. providing insufficient information or data on the scope or quality of the service provided;
  6. unduly benefitting from unilateral obligations that are not justified by the nature or the scope of the activity provided; in particular, by making the quality of the service provided conditional upon data transfer in an unnecessary or disproportionate amount”.

In sum, by overturning the economic dependence burden of proof on the platforms’ shoulders, if approved, this ambitious reform proposal will lead the enforcement toolkit under exam towards applications never tested before, opening a new chapter of the intriguing debate on the intertangles balance between “purely antitrust based” enforcement instruments and contract law grounded ones.

The second proposal sees the ICA willing to be armed with an ex-ante regulatory tool for digital platforms modelled on the recently implemented German one. Clarifying why an antitrust authority asks to be equipped such powers would need an ad hoc article (and probably a more competent author). Yet, for those brave enough desiring to learn more about the challenges competition law is facing in dealing with digital platforms, these reports are a good first step:

Herein it is merely underlined how it is widely accepted among antitrust scholars and practitioners the traditional ex-post antitrust enforcement rules are keeping pace with the challenges posed by global digital players. The abovementioned Commission report notably stressed the digitalization of the economy requires “a vigorous competition policy regime” that, nevertheless, “will require a rethinking of the tool of analysis and enforcement”. This rethinking is currently leading within the EU towards the development of a synergy between antitrust enforcement and regulatory legal instruments (this should not surprise, having Europe a long tradition of combining competition law and regulation in order to achieve similar policy objectives – e.g. the regulation of international roaming charges, Interchange Fee Regulation for the financial industry). The EU Commission draft Digital Market Act, being the last and most debated example of this policy choice, aims at monitoring certain anticompetitive behaviors carried out by the so-called Gatekeepers, those market actors that control the entry of new players into the digital market.

National Competition Authorities are pursuing the same path. In January 2021, the German Parliament adopted an amendment of the German Antitrust Act, introducing, among other things, a new weapon for the Bundeskartellamt to act against platforms and assess network effects. The reform proposal of the ICA mirrors this amendment.

This last introduces the legal notion of undertakings having “primary importance for competition in multiple markets (i.e. the Gatekeepers). The ICA shall declare this peculiar legal status following this non-exhaustive list of criteria:

  1. the dominant position in one or more markets;
  2. the vertical integration and/or presence on otherwise related markets;
  3. the access to data relevant to competition;
  4. the importance of the activities for third parties’ access to supply and sales markets and the
  5. related influence on third parties’ business activities.”

Once approved the status of Gatekeeper, in addition to the general rules governing abuse of a dominant position contained in 3 of the Law n. 287/90, the undertaking concerned is subject – for five years since the ICA declaration – to seven black-listed conducts:

  1. giving preferential treatment, in the supply and sales markets, to its own goods or services against other companies’ ones, in particular by giving preference to its own offers in the presentation or by pre-installing its own offers on devices or to integrate them in any other way into the companies’ offers;
  2. taking measures that hinder other companies in their business activities on procurement or sales markets, if the undertaking’s activities are important for access to these markets;
  3. hindering other companies on a market on which the undertaking can rapidly expand its position, even without being dominant, in particular through tying or bundling strategies;
  4. processing competitively sensitive data collected by the undertaking, to create or appreciably raise barriers to market entry;
  5. impeding the interoperability of products or services or the portability of data;
  6. providing other companies with insufficient information about the services provided or otherwise make it difficult for them to assess the value of this service;
  7. demanding advantages for the treatment of another company’s offers which are disproportionate to the reason for the demand, in particular by demanding for their presentation the transfer of data or rights which are not absolutely necessary for this purpose.

The aforementioned prohibitions do not apply to the extent that the Gatekeeper demonstrates that its conduct is objectively justified. It is not yet clear whether this burden on companies is to be understood in the same way as the traditional discipline on abuse or whether in addition to reasons of efficiency, public interest reasons (e.g. safety) could also be included. In case of non-compliance, the ICA could fine the undertakings concerned and/or impose behavioral or structural remedies to put an end to the infringement and its effects or to prevent a repeat of the conduct.

In conclusion, while the above reported reports made the last years a period of reflection for competition law, the antitrust watchdogs are now making their moves to tackle a number of shortcomings the rise of platforms led to. Despite sharing a basic premise – the perceived inherent limits of ex-post antitrust enforcement – as far as it can be assessed at this preliminary stage, the regulatory models under discussion give rise to several questions.

It is still unclear, indeed, how the concurrent application of antitrust and regulatory powers (both at EU and national level) will work when a conduct simultaneously violates competition law and harms market contestability or P2B fairness. On the same vein, how the synergies between antitrust and regulatory powers can be maximized without harming legal certainty and predictability (without forgetting procedural safeguards, above all, the ne bis in idem principle)?

One could further pinpoint an apparent lack of awareness of the intersection between competition and other regulatory protections – such as, for instance, privacy – in the digital sector, and thus the need for forms of coordination and shared responsibility among all authorities with functions in the digital sector. Indeed, some of the above reported obligations on Gatekeeper, such as those relating to interoperability and data portability and to the prohibition to introduce or reinforce entry barriers based on data exploitation, could give rise to tensions with the legislation – and the Authority empowered to surveilon the processing of personal data. In this sense, the absence of institutional cooperation let foresee a bumping road ahead for the governance of the digital economy. Yet, by providing for, as an example, rules on information exchanges between authorities or even co-responsibility between authorities in some areas, the Italian legislator could still anticipate these overlaps (not least giving the impression that taking the best from other countries should not set critical thinking aside).

Andrea Aguggia

“To Sample, Or Not To Sample, That Is The Question”

On 16th September 2020 the United States District Court for the Central District of California had to decide if the use by an artist – known as Nicky Minaj – of the recording of lyrics and melodies of a musical work “Baby Can I Hold You” by the artist Tracy Chapman (hereinafter the “Work”) for artistic experimentation and for the purpose of securing a license from the copyright owner is fair use (full decision here). Nicky Minaj was aware that she needed to obtain a license to publish a remake of the Work as her remake incorporated many lyrics and vocal melodies from the Work. Minaj made several requests to Chapman to obtain a license, but Chapman denied each request. Minaj did not include her remake of Sorry in her album. She contacted DJ Aston George Taylor, professionally known as DJ Funk Master Flex, and asked if he would preview a record that was not on her album.

The Court recognized fair use based on the following assessments:

·       the purpose of Minaj’s new work was experimentation. Since Minaj “never intended to exploit the Work without a license” and excluded the new work from her album, Minaj’s use was not purely commercial. In addition, the Court noted that “artists usually experiment with works before seeking licenses, and rights holders usually ask to see a proposed work before approving a license” The Court expressed concern that “the eradication … [these] common practices would limit creativity and stifle innovation in the music industry“;

·       the nature of the copyrighted work, did not favor fair use because the composition is a musical work, which is “the type of work that is at the core of copyright’s protective purpose“;

·       the amount of the portion used in relation to the work as a whole, favored fair use. Although Minaj’s new work incorporated many of the composition’s lyrics and vocal melodies, the material used by Minaj “was no more than necessary to show Chapman how [Minaj] intended to use the composition in the new work“;

·       the effect of the use on the potential market or value of the copyrighted Work, favored fair use because “there is no evidence that the new work usurps any potential market for Chapman“.

Considering the factors together, the Court found that Minaj’s use was fair and granted partial summary judgment in favor of Minaj that her use did not infringe Chapman’s right to create derivative works. The Court determined that Chapman’s distribution claim has to be tried and resolved by a jury, but a settlement eliminates the need for a trial. Minaj has paid a significant sum (450.000,00 Us dollar) to settle and avoid the risk of trial. If on one hand, this case confirm that private sampling should be protected as fair use, on the other hand it sounds like a warning for artists on sampling matter. Obtaining a preliminary license – also in the land of fair use – is always the best practice, although creativity and experimentation needs – in the opinion of the writer – to be protected to empower the spread of different music genre and contribute on cultural renaissance, especially regarding hip hop music, that is historically based on sampling.

The decision offers an interesting comparison with the Pelham case (CJEU – C-476/17 Pelham GmbH and others) in order to analyze how the two different systems are evolving on sampling matter.  Actually, the agreement between these two decisions is only partial.

Indeed, in Pelham the CJEU recognized the admissibility of “unrecognizable sample“. According to the CJEU “where a user, in exercising the freedom of the arts, takes a sound sample from a phonogram in order to use it, in a modified form unrecognizable to the ear, in a new work, it must be held that such use does not constitute ‘reproduction’ within the meaning of Article 2(c) of Directive 2001/29.”

Furthermore, in Pelham CJEU argue that the reproduction of a sound sample, even if very short, constitutes a reproduction that falls within the exclusive rights granted to the producer of phonogram. Considering that the US Court stressed that “not only (…) the quantity of the materials used, but about their quality and importance, too” has to be considered, according to Campbell, 510 U.S. at 587, this is probably one of the main gaps between the two decisions.

Indeed, the logical-argumentative process of the US Judge moves from a deep context analysis that implies an interpretation of sampling based on the purpose and character of the uses, according to the common-law tradition of fair use adjudication that always preferes a case-by-case analysis rather than bright-line rules.

Instead, the CJEU chose a different approach, arguing that the “free use” is a derogation not provided by the Infosoc Directive, so any reproduction act is subject to the reproduction rights notion mentioned by art.2 of each Directive. This “static” approach also (and especially?) depends on the pending – and unsolved – harmonization process of the European system of exceptions and limitations provided by the Infosoc Directive.

The US Court, instead of being based on a parameter of appreciation such as the “recognizability of hearing”, comes to the balance through an analysis of context aimed at preserving the freedom of artists to experiment, demonstrating – even in the (apparent) identity of results – more courage, as opposed to the practical approach of the European Court of Justice. The CJEU has not – in the opinion of the writer – taken the opportunity to move more decisively towards a grater balance between exclusive rights and fundamental freedoms, which should be considered the freedom to experiment for artists.

Matteo Falcolini

Chapman v. Maraj No. 2:18-cv-09088-VAP-SS (C.D. Cal. Sept. 16, 2020)

What happens when copyright protection expires? Trade mark/copyright intersection and George Orwell’s ‘Animal farm’ and ‘1984’

The 1st of January of each calendar year marks not only the days full of new years’ hope, resolutions, and promises, but also the public domain day. 1 January 2021, a moment full of hope with the anti-COVID vaccine rolling out, is no different for copyright law purposes. On the 1st of January, many copyright protected works fell in the public domain.

This year marked the falling into the public domain of the works of George Orwell. Born in 1903, under the real name of Eric Arthur Blair, he has authored masterpieces such as ‘1984’ and ‘Animal Farm’, which in recent years have become extremely topical and relevant. In early 2017, the sales of ‘1984’ went so high up that the book became a bestseller once again. Some have suggested this is a direct response to the US Presidency at the time in the face of Donald Trump.

Orwell passed away in 1950, which means that, following the life of the author plus seventy years rule as per Article 1 of the Term Directive, copyright in Orwell’s works expired on 1 January 2021. Despite this, in the last several years an interesting trend has prominently emerged. Once copyright in famous works, such as those at issue, has expired, the body managing the IPRs of the author has often sought to extend the IP protection in the titles by resorting to trade mark applications. At the EUIPO, this has been successful for ‘Le journal d’Anne Frank’ (31/08/2015, R 2401/2014-4, Le journal d’Anne Frank), but not for ‘The Jungle Book’ (18/03/2015, R 118/2014-1, THE JUNGLE BOOK) nor ‘Pinocchio’ (25/02/2015, R 1856/2013-2, PINOCCHIO).

This is the path that the ‘GEORGE ORWELL’, ‘ANIMAL FARM’ and ‘1984’ signs are now following. The question of their registrability as trade marks is currently pending before the EUIPO’s Grand Board of Appeal. This post will only focus on the literary work titles – ‘ANIMAL FARM’ and ‘1984’, as the trade mark protection of famous authors’ personal names is a minefield of its own, deserving a separate post.

The first instance refusal

In March 2018, the Estate of the Late Sonia Brownell Orwell sought to register ‘ANIMAL FARM’ and ‘1984’ as EU trade marks for various goods and services among which books, publications, digital media, recordings, games, board games, toys, as well as entertainment, cultural activities and educations services. The Estate of the Late Sonia Brownell Orwell manages the IPRs of George Orwell and is named after his second and late wife – Sonia Mary Brownwell.

The first instance refused the registration of the signs as each of these was considered a “famous title of an artistic work” and consequently “perceived by the public as such title and not as a mark indicating the origin of the goods and services at hand”. The grounds were Article 7(1)(b) and 7(2) EUTMR – lack of distinctive character.

The Boards of Appeal

The Estate was not satisfied with this result and filed an appeal. Having dealt with some preliminary issues relating to the potential link of ‘ANIMAL FARM’ to board games simulating a farm life, the Board turned to the thorny issue of registering titles of literary works as trade marks. The Board points out that while this is not the very first case of its kind, the practice in the Office and the Board of Appeal has been diverging. Some applications consisting of titles of books or of a well-known character are registered as marks since they may, even if they are well-known, still be perceived by the public also as an indicator of source for printed matter or education services. This was the case with ‘Le journal d’Anne Frank’. Other times, a famous title has been seen as information of the content or the subject matter of the goods and services being considered as non-distinctive and descriptive in the meaning of Article 7(1)(b) and (c) EUTMR. This was the case for ‘THE JUNGLE BOOK’, ‘PINOCCHIO’ and ‘WINNETOU’. The EUIPO guidelines on this matter are not entirely clear. With that mind, on 20 June 2020, the case has been referred to the Grand Board of Appeal at the EUIPO. Pursuant to Article 37(1) of Delegated Regulation 2017/1430, the Board can refer a case to the Grand Board if it observes that the Boards of Appeal have issued diverging decisions on a point of law which is liable to affect the outcome of the case. This seems to be precisely the situation here. The Grand Board has not yet taken its decision.

Comment

The EUIPO’s Grand Board is a bit like the CJEU’s Grand Chamber – cases with particular importance where no harmonised practice exists are referred to it (Article 60, Rules of Procedure, CJEU). The Grand Board has one specific feature which the other five ‘traditional’ Boards lack. Interested parties can submit written observations, the otherwise known ‘amicus curiae’ briefs (Article 23, Rules of Procedure of the EUIPO Boards of Appeal). In fact, in this specific instance, INTA has already expressed its position in support of the registration of the titles.

It must be observed here that when the trade marks were filed, back in March 2018, ‘Animal farm’ and ‘1984’ were still within copyright protection (but not for long). Indeed, the Estate underlines in one of its statements from July 2019 that “George Orwell’s work 1984 is still subject to copyright protection. The EUIPO Work Manual (ie, the EUIPO guidelines) specifically states that where copyright is still running there is a presumption of good faith and the mark should be registered”. While this is perfectly true, what is also important to mention here is that trade mark registration does not take place overnight, especially when it comes to controversial issues such as the question of registering book titles as trade marks.

The topic of extending the IP life of works in which copyright has expired has seen several other examples and often brings to the edge of their seat prominent IP professors, as well as trade mark examiners and Board members. Furthermore, several years ago, the EFTA court considered the registrability as trade marks of many visual works and sculptures of the Norwegian sculptor Gustav Vigeland (see the author’s photo below).

‘Angry Boy’ by Gustav Vigeland, Vigeland park, Oslo, photo by Alina Trapova

At stake here was a potential trade mark protection for an artistic work in which copyright had expired. One of these is the famous ‘Angry Boy’ sculpture shown here. The Municipality of the city of Oslo had sought trade mark registration of approximately 90 of Vigeland’s works. The applications were rejected. The grounds included not only the well-known descriptiveness and non-distinctiveness, but also an additional objection on the grounds of public policy and morality. Eventually, the case went all the way up to the EFTA court. The final decision concludes thatit may be contrary to public policy in certain circumstances, to proceed to register a trade mark in respect of a well-known copyright work of art, where the copyright protection in that work has expired or is about to expire. The status of that well known work of art including the cultural status in the perception of the general public for that work of art may be taken into account”. This approach does make good sense as it focuses on the specific peculiarities of copyright law (e.g., different to trade mark law, copyright protection cannot be renewed), but it also considers re-appropriation of cultural expression as aggressive techniques of artificially prolonging IP protection – something, Justice Scalia at US Supreme Court has labelled as “mutant copyright” in Dastar v. Twentieth Century Fox in 2003.

In the EU, the public policy/morality ground has been traditionally relied on to object to obscene expression. The focus has been on the whether the sign offends, thus tying morality to public policy (See also another Grand Board decision on public policy and morality: 30/01/2019, R958/2017-G, ‘BREXIT’). A notable example is the attempt to register the ‘Mona Lisa’ painting as a trade mark in Germany. The sign was eventually not registered, but not due to clash with public policy and potential artificial extension of copyright through the backdoor of trade marks, but because the sign lacked distinctiveness. Consequently, the EU understands the public policy and morality ground in a rather narrow and limited manner, namely linked only to offensive use.

Overlap of IPRs happens all the time: a patent turns into a copyright claim (C-310/17 – Levola Hengelo and C-833/18 – Brompton Bicycle), whereas design and copyright may be able to co-habit in the same item (C-683/17 – Cofemel). All would agree that an unequivocal law prohibiting overlaps of IPRs is not desirable from a policy perspective. Yet, you may feel somehow cheated when your favourite novel is finally in the public domain due to copyright expiring (which, by the way, already lasts the life of the author plus seventy years), but protection has been “revived” through a trade mark. Trade marks initially last for 10 years, but they can be renewed and thus protection lasts potentially forever (what a revival!). Therefore, there may be some very strong policy reasons against this specific type of IP overlap and extension of IPRs. Like the European Copyright Society has said in relation to Vigeland, “the registration of signs of cultural significance for products or services that are directly related to the cultural domain may seriously impede the free use of works which ought to be in the public domain. For example, the registration of the title of a book for the class of products including books, theatre plays and films, would render meaningless the freedom to use public domain works in new forms of exploitation”. To that end, Martin Senftleben’s new book turns to this tragic clash between culture and commerce. He vocally criticises the “corrosive effect of indefinitely renewable trademark rights” with regard to cultural creativity.

As for ‘Animal farm’ and ‘1984’, the EUIPO’s first instance did not go down the public policy grounds road. This is unfortunate as lack of distinctiveness for well-known titles such as ‘Animal Farm’ and ‘1984’ is difficult to articulate as the Board of Appeal itself underlines – the Office’s guidelines are confusing (some titles have been protected and others not). Besides, lack of inherent distinctiveness can be saved by virtue of acquired distinctiveness (something the Estate of the Late Sonia Brownell Orwell has explicitly mentioned they will be able to prove, should they need to). Considering that the rightholder here would be the Estate of the Orwell family, proving acquired distinctiveness of the titles for the requested goods and services would not be particularly difficult. On that note, a rejection on the ground of public policy/morality can never be remedied through acquired distinctiveness. Thus, it would perhaps have been more suitable to rely on public policy and establish that artificially extending the IP life of these cultural works is not desirable.

Well, the jury is out. One thing is for sure – the discussion at the Grand Board will be heated.

Alina Trapova

Chip patents infringement: Texas jury orders Intel to pay $2.18 billion

On March 2, 2021, a Texas jury (US District Court for the Western District of Texas, Waco), has handed down a verdict ordering Intel Corporation to pay $2.18 billion as a result of the infringement of two patents held by the company VLSI Technology LLC. (full jury’s verdict form available here).

The jury’s verdict was given against Intel after losing a patent-infringement trial over technology related to computer microprocessors (chips).

The jury, through Questions 6, 7 and 8 of the verdict amounted a lump sum of $1.5 billion for the infringement of U.S. Patent No. 7,523,373 and $675 million for the infringement of U.S. patent No. 7,725,759.

VLSI Technology affirmed that Intel was “willfully blind” of the existence of the patents in the infringement. Interestingly, while the jury found Intel guilty of such violation, it denied the existence of a “willful” infringement from Intel (Question 4). In such case, District Judge Alan Albright would have been able to raise the amount up to three times of the amount set by the jury.

A compensation of $2.18 billion is a very considerable amount, given the fact that VLSI profits are not generated from the exploitation of the patent. If endorsed, such verdict will be one of the largest patent damages given in United States history.

Alessandro Gai

US District Court for the Western District of Texas, Waco Division, jury verdict of March 2, 2021, VLSI Technology LLC v. Intel Corp, Case No. 6:21-cv-00057-ADA.




CJEU Rules: Access to Data From Electronic Communications Is Permitted Only to Combat Serious Crime and Only With Adequate Procedural Safeguards.

The Court of Justice of the European Union (“Court”) recently published a decision regarding article 15 of Directive 2002/58/EC, on privacy and electronic communications (“Directive”) stating the existence of a serious interference between data access and fundamental rights that implies the irrelevance of the amount of data accessed and the necessity for Member States to set safeguards in case such access must be nevertheless accorded to state authorities.

It must be said that the Grand Chamber substantially confirms its own case law making express reference to the judgments Tele2 Sverige (C‑203/15 and C‑698/15), La Quadrature du Net (C‑511/18, C‑512/18 and C‑520/18) but it also gives some clarifications.

The case at stake is based on a criminal conviction relying – among other things – on data generated in the context of the provision of electronic communications services. The Riigikohus (Supreme Court, Estonia), expressed doubts as to whether the conditions under which the investigating authority had access to those data were compatible with EU law – in particular article 15 of the Directive – and referred three different questions to the Court.

The first regards the relevance, in the proportionality test, of the period to which the retained data accessed by the authorities relate. The second regards the need to assess the proportionality between the amount of data to which the authorities may have access to, the seriousness of the criminal offence and the tolerated interference with fundamental rights. The third regards the quality of independence of the authority that can be competent for the prior review of the data access request. 

Regarding the first two questions, the Court holds that the Directive, read in the light of the Charter, precludes national legislation that allows public authorities to have access to traffic or location data without restrictions on the merit of the procedures, that should instead be limited to serious crimes or threats to public security. What is morethe length of the period in respect of which access to data is sought and the quantity or nature of the data available are irrelevant circumstances in the assessment of legality of the data access, being the interference with the fundamental rights in any event serious per se

With reference to the independence principle, the Court holds that that the Directive also precludes national legislation that recognises to the public prosecutor’s office the power to authorise access to data for the purpose of conducting a criminal investigation.

Indeed, even if it is for national law to determine the conditions under which access to data must be granted, the legislation must set minimum safeguards in order to avoid risk of abuse and to reconcile the various interests and rights at issue. It follows that the requirement of independence of the authority entrusted with carrying out the prior review of legitimacy of data access requests implies that the said authority must be a third party in relation to the one filing the request. This is not the case for the public prosecutor’s office when it directs the investigation and also performs the subsequent public prosecution. 

Andrea Giulia Monteleone

Case C‑746/18 – H. K. v Prokuratuur – Judgment of the Court (Grand Chamber), 2 March 2021