Monthly Archives: March 2014

10 Points on Recent EPO and UK Case Law

This is based on a talk given by Philip Webber at the CIPA Life Sciences Conference on 15 November 2013.

1.  As background to human embryo stem cells: the Biotech Directive lists inventions which are seen as immoral.  This was brought into the Rules of the EPC in 1999.  The Enlarged Board of the EPO decided in WARF G2/06 in 2008 that if an invention required a human embryo to be destroyed then it was not patentable, though it seemed from the decision that if appropriate embryo stem cell lines were publically available at the priority date then the invention was patentable, even if creation of those cell lines required destruction of an embryo.

However in 2011 the CJEU decided in C34/10 Brustle that if the invention made use of embryo stem cells that had required destruction of an embryo at any stage then it was not patentable, i.e. a stricter test than the EPO set in WARF G2/06.  In 2012 the EPO informally indicated that it would follow the CJEU’s decision.

2.  There are 3 dates which are important when looking at the patentability of embryo stem cells.  Firstly, embryo stem cell lines were first made in December 1997.  Secondly, the EPO accepts that human embryo stem cell lines were publically available from May 2003.  Thirdly, in February 2008 a paper was published showing how embryo stem cell lines could be made without destroying an embryo, essentially by taking a cell from an 8 cell embryo and letting the remaining 7 cells continue to develop.

3.  In 2013 there was a new referral to the CJEU from the UK on whether a parthenote is a human embryo (presently pending as C-364/14).  The patent application concerned activating an unfertilised oocyte and causing it to develop into a parthenote.  Embryo stem cells can be produced from the resultant parthenote, but the parthenote is incapable of developing into a human.  The Advocate General’s opinion in the Brustle decision had been in favour of a test requiring the embryo to be capable of forming a viable human, but the subsequent CJEU decision did not include this.

4.  The German Brustle patent has now been granted with a disclaimer to embryonic stem cell lines which are derived by destruction of a human embryo.  There is a corresponding European patent which has the same disclaimer and which has been opposed by Geron on the grounds of morality and insufficiency.  Geron argued that at the priority date it was not possible to make embryo stem cell lines without destroying an embryo.  The opposition division revoked the patent in June 13 on the basis that the sufficiency issue led to the disclaimer adding matter.  This will now go to appeal.

It is clear from European Patent No. 2283117 granted in October 2013 that the EPO is granting cases filed after 2008 without an appropriate disclaimer being required.

5.  In the UK High Court , Lilly v Janssen concerned treating Alzheimer’s with an antibody to amyloid-β.  The prior art, Konig, disclosed the concept of using such antibodies to treat Alzheimer’s but did not provide any data.  The question was whether this disclosure was novelty-destroying.  The UKIPO Guidelines on medical use claims say that there does not need to be disclosure of clinical trials in a prior art document for it to be novelty destroying, but there is no further clarification of what sort of disclosure would be novelty-destroying.  The EPO decision T1437/07 says the sufficiency test must be identical for the prior art and for the patent, and thus there needs to be credible disclosure of the treatment.  A previous UK House of Lords decision, Synthon v SmithKline Beecham had said that anticipation and enablement must be same test.  Ultimately the UK Court decided in Lilly v Janssen that since Konig had not have experimental data it is speculative, and therefore not novelty destroying.  Given there were reasons to expect the antibodies would not work the claims were also inventive over Konig.

6.  The Court in Lilly v Janssen also considered sufficiency, both classical and Biogen insufficiency (the latter corresponds to the EPO’s Agrevo test).  For Biogen insufficiency the court asked ‘do all antibodies against amyloid-β work?’  Given that only antibodies to the N terminus of the protein would work the specification did not make it plausible that any antibody against amyloid-β could be used and therefore the patent was insufficient.  The Court also considered classical insufficiency asking whether it was undue burden to make therapeutic antibodies.  The sequence of the antibodies or deposits of them were not available.  The evidence showed they would require 30 months to make.  However there were doubts about efficacy in mice and human trials.  There had been success in a phase II trial, but failure in a phase III trial.  Given all of this it was decided the patent was also insufficient for classical sufficiency.

7.  The UK IPO case Aueon (BL O/248/13) concerned a method of diagnosing cancer that included a  step of checking a database to identify potential treatment options.  The applicant argued that one can have technical and nontechnical steps in a diagnostic claim given the EPO decision G1/04.  However the hearing officer treated the invention as computer program and used the approach in Aerotel, leading to refusal.

8.  The UK IPO case Lalvani (BL O/220/13) concerned inventive step of a formulation for lactating mothers.  Essentially the hearing officer said that all of the ingredients recited in the claim would be obvious to use in the formulation, and in the absence of showing synergy the invention was obvious.  The applicant’s arguments that there was bound to be synergy in such a biological system was rejected in the absence of data.

9.  The UK High Court case Glenmark v Wellcome concerned the obviousness of an antimalarial with two components in in a particular ratio.  The Court said that if there was an urgent need the skilled person would pursue a ‘hope to succeed’ which is different from the usual test of reasonable expectation.  The claims were found to be obvious in view of a prior art oral presentation.

10. There is the perception that the EPO is getting stricter every year on added matter.  However recent decisions T667/08 and T2619/11 may make it easier for applicants.  Essentially these require the EPO to look at the application in its entirety, and not just at the literal wording.  The EPO Guidelines have been changed as a result to make it clear that implicit features should be taken into account when judging added matter.

You may also wish to see related articles Top 10 Points from EPO Case Law in 2013 and Top 10 Points from IPO Decisions in 2013.

10 Points on US Patent Law Developments Focusing on Biotech

This is based on a talk given by David Resnick at the CIPA Life Sciences Conference on 15 November 2013.  Points 1 to 6 relate to patent eligible matter.  Points 7 to 10 relate to general patent matters.

1.  The most important developments have been in the area of patent eligible subject matter.  Section 101 is now an important ‘gatekeeper’ being used by the USPTO and the Courts to decide on what can be patented.  Essentially laws of nature and abstract ideas cannot be patented.  Recent cases have overturned settled practice in the areas of diagnostic tests and compositions of biological matter.

2.  The Labcorp, Bilski, Prometheus and Myriad cases have changed the situation dramatically for judging patent eligible matter.  The US Supreme Court decided in Labcorp v Metabolite that a claim to assaying body fluid for diagnosis was not patentable because this was a natural phenomenon.  The Federal Circuit introduced the ‘machine or transformation’ test in Bilski to determine what was patentable in such a situation, but the Supreme Court on the same case said this was not the only test, without actually defining what the test should be.  In Prometheus again a diagnostic test relating to measuring the level of a substance in the body was held to be unpatentable.  These three cases have left a lot of uncertainty as to the boundaries of eligible subject matter in the diagnostic area.

3.  Myriad held that DNA isolated from the natural state is not eligible matter.  However changing it to cDNA makes it different from nature and so cDNA is eligible matter even though the sequence of cDNA is dictated by nature.  This has led to uncertainty for other important biological materials, such as antibiotics, enzymes, isolated cells, and protein drugs.

4.  In order to overcome section 101 objections against product claims to it may be helpful to add the features of being ‘synthetic’ or ‘recombinant’.  Pairs of primers might be patentable.  Chemically modified versions of naturally occurring DNA might be patentable.

For diagnostic claims the following may be helpful

– adding a treatment step

– limiting to testing a subpopulation (e.g. a person with a particular symptom)

– using the word ‘measure’ instead of ‘determine’

– testing multiple markers

– referring to an ‘assay comprising’

– specifying the test in more detail, e.g. allowing antibody to bind, washing… etc.

– specifying a type of sample or timepoint at which it was taken.

5.  The Courts are looking at ‘law of nature’ very broadly.  In Perkin Elmer v Intema the Federal Circuit rejected as patent ineligible a test which compared first and second trimester samples to detect the likelihood of the foetus having Down’s syndrome.  In Ariosa v Sequenom a test looking at nuclei acids from a foetus in maternal blood was held to not be eligible matter.  However Myriad is presently in litigation with several parties and that should give more clarity to the law in this area.

6.  In regards to strategy for applicants, for issued patents it is probably not a good idea to go for reissue to rewrite claims in view of the uncertainty in the area.  For pending applications it would be a good idea to limit claims to make them more specific, perhaps focusing on the specific product at point of care that needs to be protected.  When filing new applications it might be a good idea review what information might be kept as a trade secret.  In addition it is worth keeping an eye on trends in what claims the USPTO is allowing to proceed to grant.

7.  It must be remembered that anything said in prosecution can be used in a claim construction proceeding.  In Biogen v GSK, Biogen failing to challenge the characterisation of the invention by the Examiner was important in claim construction.  Responding to a restriction requirement can also lead to arguments being on file that can be used against you in subsequent proceedings, as happened in Uship v United States.  It may be a better strategy to either not respond or respond without using technical arguments.

In Teva v Sandoz the average molecular weight could be measured in different ways and so the Federal Circuit invalidated certain claims as insolubly ambiguous.  Across a portfolio one needs to have a consistent approach, and perhaps also use different terms in different cases.

8.  In Hamilton Beach Brands v Sunbeam it was found surprisingly that the on-sale bar under pre-AIA rules applied even in the case of a supplier selling ‘privately’ to the company that had developed the product.  This could be different under AIA because an offer to sell also has to be ‘available to public’ under the new rules.

9.  It must be borne in mind that there are circumstances where old cases will be treated under the new law.  In particular if new claims are added which are not entitled to a pre-AIA filing date.  If such claims are introduced when bringing an application into the US national phase then it is something that cannot be fixed, and so it is better to amend claims later as a preliminary amendment.

10.  Whilst it is possible to speed up examination using prioritised examination, in practice the requirements on the applicant are onerous, and it is better to use PPH instead if possible.

You may also be interested in the related articles Top 10 Points on Patent Developments and Case Law in the US in 2013 and Top 10 Observations on US Restriction Requirements.

10 Points on the European Unitary Patent and the Unified Patent Court

This is based on a talk given by Kevin Mooney to the CIPA Life Sciences Conference on 14 November 2013.

1.  The European unitary patent has come about in response to the perception that diverse litigation systems across Europe are a problem.  There are significant difference in procedure (e.g. the extent of reliance on expert witnesses) and in the speed and costs of the different systems.   The quest for unitary patent system started in 1975 with the Community Patent Convention.

2.  The objectives of the new system are a single patent, a single litigation procedure with experienced judges, quality, efficiency and legal certainty.  One incentive to use the new procedure will probably be reduced costs due to reduced renewal fees, although this is not straightforward as under the present ‘bundle of patents’ system renewal costs can be decreased in later years by reduction of designations (which won’t be possible with the unitary patent).

3.  The Unified Patent Court (UPC) will have local, regional and central divisions.  This gives a plaintiff lots of places to sue.  The language will also be chosen by the plaintiff though there provisions for language to be the one in which the defendant normally conducts business.

4.  We are now at the 15th draft of the rules of procedure [In fact subsequent to this talk we are now at the 16th draft].  The litigation procedure has 3 parts:

– a written part lasting 6 months  (claims, counterclaims, applications to amend the patent, etc)

– an interim procedure lasting 3 months, where the judge-rapporteur is essentially preparing the case for a oral hearing

– the oral hearing, typically lasting 1 day.

5.  The appeal will not be a retrial and normally no new evidence will be submitted.

6.  The Court will have a broad discretion as to whether to bifurcate at end of the written procedure. Bifurcation is expected to be rare.  The Court will have the discretion to stay infringement proceedings if it decides to bifurcate and the expectation is that they would stay.

7.  The conditions for provisional measures (injunctions, etc) are that the applicant is entitled to commence proceedings, the patent is valid and the right has been infringed.  In addition the Court must weigh up the interests of parties.  There is also a provision for compensation of defendants if the provisional measure was incorrectly granted.

8.  At the moment there is discussion of whether to charge for opting out of the unitary patent (back to the present ‘bundle of patent’ system).  The pharma and biotech sectors have indicated they would opt out since the new system is untried.

9.  For the litigation there is a fixed fee for each part, and a controversial value-based fee for counterclaims for revocation.

10.  Representation is a much discussed point.  Lawyers or patent attorney will have representation rights and these are being defined very broadly.

You may also wish to see related articles 10 Observations on Patent Litigation and 8 Disadvantages of a Specialist Patent Court.

10 Points on Research and Bolar Exemptions

This is based on a talk by Graham Stuart given to the CIPA Life Sciences Conference on 14 November 2013.  The talk was essentially focused on the position of the exemptions in the UK.

1.  When analysing whether something falls within the research exemption one should determine whether one is working ‘on’ the invention or ‘with’ the invention.  So generally if one is looking at the underlying principles of a patent and building on it then that should fall in the exemption.  However in the case of research tools one will be using them (working with them) in one’s research, rather than seeking to study them or build on them, and so they will not fall within the research exemption.

2.  Developing a drug requires many different activities that will fall within the research exemption. The exploratory and drug discovery phase should fall within the research exemption.  Essentially making, testing and validating patented subject matter should be exempt.

3.  However Inhale v Quadrant confirmed that supplying the means to make and test patented subject matter did not fall within the exemption when this was done by another company.   That means smaller companies that cannot themselves make the patented products are at a disadvantage because they cannot outsource this.  However a third party contracted to do the experimental use is considered exempted because it is viewed as a co-experimenter.

4.  In the case of research tools it is generally difficult to argue that the exemption applies.  In general large companies with patents on research tools may find that they are best used to obtain injunctions against others.  In Switzerland and Belgium though there are provisions which make it easier to use patented research tools.

5.  Testing candidate drugs in preclinical studies will probably be exempt given that one is trying to find out something new (Monsanto v Stauffer).  Clinical trials would also usually be exempted.

6.  The Bolar exemption covers work carried out to obtain regulatory approval for a generic drug.  In the EU this came into being through a Directive. It has been incorporated into national provisions either broadly or narrowly. ‘Broadly’ means work to obtain any marketing approval is covered, whilst ‘narrowly’ means only work relating to approval of a generic drug is covered.  The UK has a narrow provision, and at the moment it seems the Unified Patent Court will adopt a narrow provision.

7.  The activities which are allowable under the Bolar exemption in the UK are well established, i.e. carrying out tests, manufacturing enough to test, developing the pharmaceutical composition, etc. The German Courts decided in Astellas v Polpharma that supplying of a substance was not covered by the Bolar exemption.

8.  In the US Merck v Integra determines the extent to which work is covered by the safe harbour provision, which concerns generating data for submission to the FDA.

9.  There is uncertainty in many territories over the extent to which research tools are covered by the Bolar exemption.

10.  One hot area which may lead to these provisions being tested in litigation is biomarkers, i.e. personalised medicines.  The use of biomarkers is often covered by layers of patents, and companies will want to test to what extent exemptions apply to testing biomarkers.  In fact it may be in the interests of companies that own patents to biomarkers to let them be used during clinical trials so that they become a companion diagnostic to the drug.

You may also wish to see related articles 10 Points on the Pharma/Biotech Sector and 10 Observations on Different Types of Research.

10 Points on SPC’s (Supplementary Protection Certificates)

This is based on a talk given by Beatriz San Martin at the CIPA Life Science Conference on 14 November 2013.  Her talk had the title ‘SPC’s – Are they Fit for Purpose?’

1.  The SPC Regulation have been around for 20 years, but a lot of uncertainties remain.  Given that there is a strong commercial incentive for innovators and competitors to challenge the Regulation, purposive construction has been exploited to interpret it in different ways.  The Regulation is drafted vaguely to allow for new technologies which emerge subsequently, but that has led to a lot of uncertainty and a lot of referrals to the CJEU.  Unfortunately the CJEU is guilty of not providing clear decisions and national judges are now reluctant to make decisions based on the Regulation if any degree of interpretation is required.

The SPC Regulation works well for traditional medicines, i.e. those with single active ingredients, where it is the first marketing authorisation (MA) and it is covered by basic patent.  However, uncertainty quickly arises for situations beyond this, e.g. combination products or where the product is being used for another therapeutic indication.

2.  Most people assume that the objective of SPC’s is to compensate for patent protection which is lost due to regulatory delay.  However the explanatory memorandums and recitals for the SPC regulations for pharmaceutical and plant products do not refer to this.  Instead the objectives are stated to be protection for high levels of investment, to maintain high levels of research in these areas and to maintain competitiveness.

3.  What is the product?  Article 3 of the Regulation refers to a product which is protected by a basic patent and which is the subject of an MA and SPC.  The assumption would be that it is the same  product in these three cases.  Article 1 defines the product as an active ingredient or a combination of active ingredients.  The explanatory memorandums and recitals provide guidance and suggest a strict definition, so that different salts or use of different excipients could not lead to a subsequent new SPC.

4.  The question of what the product can be was the subject of GSK C-210/13 concerning whether an SPC was possible to an adjuvant on its own and or in combination with a flu vaccine.  Judge Arnold from the UK referred this to the CJEU because the law was not clear and because there were diverging decisions across the EU.  The essential question was ‘can an adjuvant be an active ingredient?’.  The Neurim decision had previously caused a lot of uncertainty as to defining a product.  Before Neurim, MIT C-431/04 had established a strict approach where an excipient was not considered an active ingredient.  However Neurim used a purposive construction of the Regulation [Subsequent to this talk the CJEU now said decided in GSK C-201/13 that an adjuvant is not an active ingredient because it does not give a therapeutic effect when used on its own]

5.  There is a similar situation in Bayer Cropscience C-11/13 for a plant protection product.  The decision concerns lsoxadifen which is a safener used to prevent the harmful action of herbicides.  The question is whether a safener can be an active substance.  [Subsequent to the talk the Advocate General has given the opinion that a safener can be an active substance for the purpose of the SPC Regulation, but the CJEU has yet to decide]

6.  Another issue concerns the relationship between the product and the basic patent.  In Medeva C-322/10 it was held that the SPC product had to be specified/identified in the wording of the claims of the basic patent.  In that case the patent claim referred to a product with 2 active ingredients, but the product in the MA and SPC had in addition to those 2 ingredients other active ingredients.  Thus the SPC could not rely on that patent.  The UK Court of Appeal criticised the CJEU’s decision as not providing proper guidance and being unclear.  The Medeva decision led to uncertainty in national courts and notably in the case of the Irbesarten and HCTZ combination litigation it led to different decisions across the EU.

7.  In Actavis v Sanofi C-443/12 the patent claims refer to Irbesarten and a diuretic, but not specifically to HCTZ, which is specified as the diuretic in the SPC. [Subsequent to the talk the CJEU decided that since there is already an SPC for the main product a subsequent SPC is not possible for the combination]

In Eli Lilly v HGS C-493/12 the question was whether an antibody can be sufficiently identified by simply referring to what it binds, or whether the structure is needed. [Subsequent to the talk the CJEU has now said that a structure does not need to be given for an antibody, and it can be defined just by reference to what it binds]

8.  In a pending referral Actavis v Boehringer C-577/13 the question is whether a patent can be amended post-grant to solve the issue of mentioning a specification combination product in the claims.

9.  Georgetown C-422/10 concerns how many SPC’s are possible per patent. [Subsequent to the talk the CJEU has decided that it is possible to have more than one SPC per patent where the patent protects more than one product].  A previous decision on this point, Biogen C-181/95, could be interpreted both ways.

In Neurim C-130/11 the CJEU surprisingly said that a subsequent SPC was possible because the earlier MA did not fall within the scope of protection of the later patent.

10.  Is it time for reform of the SPC system given new technologies such as biologics and gene therapy, and given the increased cost of developing medicines? The European Commission has indicated to the CJEU that reform of the SPC system is on the political agenda.

You may also wish to see related articles 10 Points on Regulatory Data Exclusivity and Advice for Scientists Setting up a Company.

10 Points on Regulatory Data Exclusivity (RDE)

This is based on the second half of a talk given by Neil Jenkins at the CIPA Life Sciences Conference on 14 November 2013.  The first part about UK Pharma Patent Cases in 2013 can be accessed here.

1.  Regulatory data exclusivity relates to marketing authorisations for pharmaceutical products. When applying for marketing authorisation to release a pharmaceutical product on to the market the relevant company needs to file a regulatory dossier application which provides clinical trial data to the relevant regulatory authority.  Such data is expensive to produce and therefore a company would want to keep it confidential.  However there is a public policy to cut down on unnecessary experiments.  In addition governments wish to encourage competition to drive down health costs.  Thus third parties can rely in part on existing applications for marketing authorisations when trying to obtain authorisations (by filing ‘bridged’ applications), and RDE is a way of dealing with the competing interests in this situation.

2.  RDE is essentially a quasi IP right which gives a company a period of exclusivity in a market in respect of competitor products which have referred to the company’s regulatory dossier application when applying for their own marketing authorisation.   TRIPS mentions RDE and says that companies shall be protected against unfair commercial uses of their data.  The right exists in many territories, though the periods vary tremendously.  In the US it links to the patent system through the orange book.  Whilst patents protect inventions, RDE protects the investment made in generating data and is an automatic right which comes into being as soon as an application for marketing authorisation is filed.

3.  In Europe RDE is governed by Directive 2001/83 as amended in 2004 and a Regulation which issued in 2004. The Directive sets out the framework for RDE whilst the regulation deals with RDE obtained via the European Medicines Agency (EMA).  Marketing authorisations can be obtained from a single national authority or from a number of countries using an ‘EPO-type’ procedure in a single application where national authorities will coordinate.  In addition marketing authorisations can be obtained centrally from the EMA.

4.  Before 2004 under the old law there were a series of cases before the ECJ which concerned line extensions of the right.  The question was if subsequent data became available for new pharmaceutical forms of the active substance would these new forms have their own RDE period?  The ECJ decided ‘no’.  The overhaul of the law in 2004 concerned 3 concepts listed below and discussed in points 5 to 7 below:

– the definition of the generic medicinal product

– the global marketing authorisation concept

– the 8+2+1 rule (8 years of data exclusivity, 2 years of market exclusivity and 1 additional year of market exclusivity for new therapeutic indications).

5.  The definition of the medicinal product was broadened to include salts and derivatives, such as immediate release forms.

6.  It was made clear that for a ‘global marketing authorisation’ where additional data is filed for new indications the new data does not have its own RDE period.  As a consequence present litigation is mainly to do with the scope of global marketing authorisations.

7.  A bridged application that refers to an earlier application can only be filed 8 years after the earlier application, and marketing authorisation will not be granted until 10 years from the earlier application, which is extended to 11 years where the earlier application refers to new indications.

8.  Enantiomers will normally be seen as derivatives of the original product and so covered by the same global marketing authorisation.  However in the case of escitalopram it was not included in the original authorisation because its ‘profile’ was sufficiently different from the original substance.  The issue was also raised in the Sepracor case.  The EMA has published a paper on this saying it essentially depends on whether the enantiomers show different properties.

9.  Combination products are dealt with in Article 10 of the Directive.  Essentially new data is needed for the combination and one cannot rely on data for the individual substances.  Whilst it was assumed that a combination would not form part of the earlier global marketing authorisation (and thus would have its own period), this has been challenged in the pending case Teva v EMA.

10.  Crossover is possible between the centralised EMA route and national routes.  The centralised route is restricted and normally needs a new active substance for a full regulatory dossier and a bridged application is only possible if the original case was EMA authorised.  There is uncertainty over the RDE period for a centralised authorisation if there have been previous national authorisations.  Is it part of the same global marketing authorisation?  This issue is pending in Novartis v Commission.

In conclusion RDE is powerful tool, giving an automatic right which has a long duration.  However there are still uncertainties as to scope of a global marketing authorisation.

You may also wish to see related articles 10 Points About Data in Patent Applications and Patent Advice for Research Companies.

10 Observations on the Success and Failings of University Tech Transfer

Universities around the world are being encouraged to find ways of commercialising their research, the assumption being that this always a good thing.  However not all university tech transfer offices are able to operate at a profit, and so questions need to be asked about whether all universities should attempt commercialisation.  The following points are from the report ‘University Start-Ups: Critical for Improving Technology Transfer’ which concerns university tech transfer in the US.

1.  All 206 US universities which carry out high levels of research activity have technology transfer offices (TTOs).  However TTOs are costly and in 2012 84% of them did not make enough money to cover staff wages and patents costs.

2.  There is great asymmetry in the distribution of licencing income for universities. The top 5% took 50% of the licensing income and the top 10% took nearly three-quarters.  This asymmetry is unchanging with only 37 universities able to reach the top 20 of licensing revenue in the last decade.

3.  TTOs often look to produce blockbuster patents, such as the gene splicing Cohen-Boyer patent that yielded $255 million for Stanford and UCSF.  However in reality such patents are rare.

4.  Universities are therefore increasingly turning to the ‘start-up’ model as a way of commercialising research.  TTOs can afford to set up start-ups with help from government and investors.  Start-ups will licence high-risk low-fee patents and provide an environment for the technology to be developed with the help of university scientists.

5.  Setting up start-ups is seen by the government as fostering entrepreneurship and attracting high-tech industry to the university’s region.

6.  For TTOs start-ups allow managing of financial risk, diversification and active management of the investment portfolio. Start-ups allow better integration of the university into the market system and more proactive partnering with the private sector.

7.  The government should provide greater experimental use exemptions for start-ups.

8.  Recognising the present asymmetry of success in commercialising research the government should seek to distribute its funding more equitably.

9.  The much-talked about ‘innovation deficit’ could be met with more imaginative ways of harnessing the new knowledge and know-how which is being produced (rather than increased government funding for research).  Start-ups may assist this.

10.  Start-ups may also be an environment which more effectively converts intellectual value to economic value (compared to university research or licencing).

The report itself can be accessed here.

You may also wish to see related articles Problems of Patenting and Commercialising University Research and Patent Advice for Tech Transfer Offices.