Selection invention is not separately defined in the Indian patent law. Selection patents are not limited to chemical and pharmaceutical field only, they are found in other technological areas, such as engineering, biotechnology, material science and telecommunications. In manufacturing processes, where operating conditions such as temperature and pressure within a known range produces unexpected effects in the known processes or the enhanced properties of the resulting products selection patents are preferred. However, to be patentable it must satisfy conditions such new, non-obvious or inventive and sufficiently disclosed. A selection patent though no different from traditional patent but these patents can cause difficulties, both in terms of their patentability and their enforceability. In some jurisdictions the judicial precedents place selection patents in a special case to "normal" rules on patentability. Nonetheless, the same principles of patentability assessments would still apply. No doubt some difficulties may arise in applying these principles when the prior-disclosed group and the selection are similar in terms of how they are named or defined. Similarly the difficulty in infringement suits would arise in same manner as would arises with patents claiming new uses for old products. A difficulty in defending selection patents invalidity challenge may arise on how the patentee chooses to craft his selection, both in terms of defining the selection from the previously known /disclosed wider class and in terms of defining the inventive feature of the selection.

Selection inventions and patent law

Inventions where with the selection of one or more specific embodiments from a general disclosure in the prior art, a new invention is carved out are known as selection inventions. Normally a general disclosure in the prior art is not considered fatal to the novelty of a later invention which claims a specific embodiment. If a prior art discloses metal, a later claim to specific metal such as nickel would still be considered novel. In fact the selection of specific dimensions, ranges of values or parameters within known larger values based on new characteristics, surprising results and unknown properties are considered to be invention within the meaning of section 2(1)(J). But the criterion for the assessment of novelty of a selection invention is not straightforward as it appears to be. It becomes more taxing when such prior art relates to lists or ranges. In India, case law concerning the novelty of selection inventions relating to lists or ranges has not evolved over time as it has been in other jurisdictions.

Selection patent: finding the unexpected /surprising results

A selection patent is granted for making an inventive choice of selection from a prior known field. Selection inventions normally involve the selection of individual elements, subsets, or sub-ranges, which have not been specifically disclosed previously, within a larger known set or range. For example, a selection patent may involve a claim to a particular group of compounds having certain advantageous properties, where that group is selected from a prior-disclosed wide class of compounds, where that advantageous property is not possessed by the prior-disclosed wide class of compounds. The principle laid down in selection patent can be deduced from the classic espousal in the case of IG Farbenindustrie (1930) 47 RPC 289 on doctrine of allowing patents to old or obvious inventions where the invention lay in the selection from a class. Justice Maugham J. in that case, worked out a three-step test for a selection patent to be valid:

  1. the patent must be based on a substantial advantage secured (or disadvantage avoided) by the selection,
  2. the whole of the selected members must possess the advantage, and
  3. the advantage of the selection must essentially be peculiar to the selected group.

Numerical ranges and selection patents

If the prior art is a numerical range for which only the end points are disclosed it is normally considered that values between the end points are not specifically disclosed. For example if an alkyl group with a chain length from one to four carbons a chain from C1 to C4 disclosed does that mean a C2 chain is also disclosed? IPO view on this aspect is neither clear from the statute nor the new guidelines specifically consider this situation. This leaves IPO examiners to look for jurisprudence elsewhere. As per EPO guidelines a C2-chain is not considered disclosed, leaving sufficient scope for choosing advantageous selection amongst the disclosed range.

Novelty determination Practice and interpretation

In India lack of novelty is determined by factors such as prior publication, public knowledge and public use and commercialised products. It is referred to as 'anticipation'. While the word 'anticipation' is not expressly defined in the Patent Act, its determination is guided by Sections 29 to 34 identify what anticipation not anticipation. There no provision in Act as such to treat selection patent as anticipated. However, in practice if a prior art describes something falling within the scope of an alleged claim then, if by studying the prior art the claimed invention can be performed, the claim would be considered to be anticipated. In India, a patent application is considered to be anticipated if the invention is disclosed in a patent or any other document which is published before the priority date of the application. The concept for identifying prior publication was established in Farbewerke Hoechst Aktiengesellschaft Vormals Meister Lucius v Unichem Laboratories, [1969 Bom 255] which is as follows:

"to anticipate a patent, a prior publication or activity must contain the whole of the invention impugned; i.e., all the features by which the particular claim attacked is limited. In other words, the anticipation must be such as to describe, or be an infringement of the claim attacked."

Similarly  in Lallubhai Chakubhai Jariwala v Chimanlal Chunilal and Co the court ruled that :

"the two features necessary to the validity of a patent are novelty and utility, but the real test is the novelty of the invention. Novelty is essential, for otherwise there would be no benefit given to the public and consequently no consideration moving from the patentee.

Therefore, factors considered for anticipation assessment of the selection invention cases are same as in other inventions. The Indian Patent Office (IPO) practice does not put bar on patentability of selection invention but the selection patents are put the test of combined reading of Sections 3(d) and 2(1)(ja) which appear to take standards regarding patentability of selection invention on a higher pedestal .Though not mandatory, Indian patent office normally follows UK practice in selection patent cases. This perhaps results from the fact that the Indian patent law derives its roots from the British law in general.

IPAB on selection patents – [Imatinib mesylate case]

In the popular Novartis (Glivec Case) IPAB made a very pertinent observation relating to the issue of selection patent :-

"IV. Selection Patent

"What we observe is that there is no reference of "selection patent" as such in the Indian patent law i.e. the Act. The patentability of an alleged invention is basically determined by establishment of novelty (anticipation), inventive step and industrial applicability of a product or a process [section2 (1)(j), and 2 (1) (l) of the Act] to the exclusion of inventions which are not patentable listed in section 3 and 4 of the Act. However, in our mind we cannot totally deny that there cannot be any possibility under the Indian law, where the required conditions as above cannot be fulfilled, for the grant of a patent in India where the inventive step is demonstrated by way of an inventive selection."

IPAB went on further to elaborated the selection invention and observed that

"we understand that the applicant for patent asking for a selection patent has to demonstrate the inventive step by way of "inventive selection" from a field that is in general terms, already known. Further, "In the context of a selection patent the inventive step will generally lie in making the discovery that what has been selected provides a genuine advantage over the generality from which it was selected. The advantage being one that could not be predicted".

IPAB even quoted the portion of the judgement of Maugham which states that

"This case seems to be the first which has arisen in these Courts in which the question of the validity of a chemical selection patent has been directly considered. It may be observed that chemical patents in recent years have consisted of two sharply divided classes. The first class is that of patents based on what may be described as a originating invention, that is, the discovery of a new reaction or a new compound. Such patents may be called for brevity "originating patents". The second class comprises patents (the so-called selection patent) based on a selection of related compounds such as the homologues and substitution derivatives of the original compounds which presumably have been described in general terms and claimed in the originating patent. The number of combinations possible is really surprising. ........... In 1897 it was estimated by Burlow in his "azofarbstoffe" that from the amines and phenols then known some 23,000 simple azo dyes of this type were possible, and, if dyes containing two azo groups were included, the number rose to over three million. It is said that over ten million are now known to be capable of manufacture. It is evident that theinventive step involved in the originating patent, forexample, such a step as the Griess reaction to which I have above referred, differs in kind from the systematic investigation or research required to ascertain that some of the combinations possible underthe originating patent,for example selectedarylides and diazo compounds madeaccording to the prior specifications, possess distinctive and itmay be unexpected properties. The question has been raised in the present case whether it is possible to show subject-matter in respect of a selection patent in the sense in which I use the word. I have come to theconclusion that sucha patent may wellbe validand that properly considered.There is no more difficulty in such a case in establishing subject matter than there is, say, in a mechanical or a combination patent. It must be remembered, of course, that the selected compounds have not been made before, or the patent would fail for want of novelty. If the selected compounds, being novel, possess a special property of an unexpected character, for example if a mono-azo dye were to be made by selecting components not hitherto employed which resulted for the first time in a green dye, I cannot see that the inventive step essentially differs from the step involved in producing a new result by a new combination of well-known parts or indeed from using the common and well-known factors (cranks, rods, toothed wheels and so forth) employed in mechanics in the construction of a new machine."

The IPAB observed that the article referred in this connection also made three general propositions relating to the selection patent :-

First, a selection patent to be valid must be based on substantial advantage (which includes avoiding substantial disadvantages) to be secured by the use of the selected members;

second, the whole of the selected members must possess the advantage in question;

Third, the selection must be in respect of a quality of a special character which can fairly be said to be peculiar to the selected group.

The said article according to the IPAB also adds how to draft a specification in case of a selection patent.

"In case of selection patent, the essence of the inventive step, that is necessary for the patentee to define in clear terms the nature of the characteristics which he alleges to be possessed by the selection for which he claims a monopoly. He has in truth disclosed no invention whatever if he merely says that he selected group possesses advantages. That apart altogether from the question of what is called sufficiency, he must disclose an invention; he fails to do this in the case of a selection for special characteristics, if he does not adequately define them."

IPAB also refered to a paragraph in page 437 of the book by P. Narayanan on "Patent Law" (4th Ed.) wherein a reference is made to the Ld. Judge's observation in Beecham Group Ltd.'s (Amoxycillin) Appln.[1980] RPC 261 at 292 (Buckley,LJ) that the principle of selection must equally apply where the alleged invention relates to only a single member of a known series of family. The said passage is given as follows:

"Although these propositions were enunciated in relation to a patent for a selected group of members of a known series or family of substances, the principle must apply equally where the alleged invention relates to only a single member of a known series of family. The alleged invention must be based on some substantial advantage to be gained from the use of that selected member and must be peculiar to it. The substance must however be truly new and the advantage to be gained from its selection must be the inventor's own discovery, as opposed to mere verification by him of previous predictions or of what was previously predictable; in other words, it must be unexpected. The selected member must be novel and must possess a special property of an unexpected character which must be useful."

In concluding remarks IPAB observed that

"we noticed particularly in chemical patents the concept of "selection patent" where the inventive step (also novelty) is demonstrated by way of an inventive selection of even a new, unexpected or unpredictable single member having surprisinglyadvantageous properties previously not known from a known series of a family disclosed in the art can be accepted in the Indianlawalso. [Emphasis added]

IPAB test of selection invention patentability

After considering the available cases laws and other related articles on selection patents IPAB shortlisted and summarised the following minimum requirements for determination of the selection inventions:-

"(1) Whether there is any statement in the specification where the nature of the invention concerns with some kind of selection.

(2) Whether the selection is from a class of substances which is already generally known.

(3) Whether the selected substance is new.

(4) Whether the selection is a result of any research by human intervention and ingenuity opposed to mere verifications.

(5) Whether the selection is unexpected or unpredictable.

(6 ) Whether the selected substance possesses any unexpected and advantageous property. "

IPAB not only laid down the six prong test to determine the selection invention but also applied this test in the present case. IPAB found that there is a statement in the specification and therefore the first condition of the test was met. The said statement described the surprised finding in respect of β-crystal form of imatinib mesylate, and which has very advantageous properties. IPAB found all the conditions laid above to meet which according to IPAB met the requirement of selection patents. However IPAB further observed that

"But for determining patentability of pharmaceutical substances mere meeting of novelty and inventive step (and industrial applicability) criteria does not entitle one to get a product patent. It has to satisfy the requirement of at least section 3(d) of the Act which says that a new salt forms, polymorphs etc. or derivatives of a known substance is not patentable unless this form demonstrates significant enhancement of properties withregard to efficacy."

From the above IPAB decision one can conclude that the selection invention must satisfy the six test requirements mention ante for further consideration for patentability under section 3 (d).In the instant case the IPAB  found that the criteria of selection invention was satisfied but the invention was not patentable under section 3(d).

Selection patent and Infringement

Court Action involving possible infringement of selection patents is yet to be seen in India. However, in order to rule for the existence of infringement, infringement of the elements of at least one independent claim of the patent in question must occur. The 'doctrine of equivalents' could also a decisive factor. Notwithstanding the fact that the selection invention is based on a new form and in order to infringe a given selection patent the infringer would have to explicitly use the new form of the invention. In other words for the infringement of the new form selection patent it is necessary to use it intentionally for the same purpose. It may not be surprising to find that the manufacturer of a product will not infringe a new form selection invention unless he practices or contributes to the practice of the invention for that particular purpose.


The Indian Patent Office (IPO) practice does not put bar on patentability of selection invention but the selection patents are put the test of combined reading of Sections 3(d) and 2(1)(ja). IPAB suggested and observed in Novartis case that selection patent is an integral concept of Indian patent law. The six step test laid by IPAB paved way for finding the existence of selection invention. No doubt selection invention finds strong presence in pharmaceutical R&D activities in identifying the lead compound millions from structural variations from a genus (base) compound. In fact the jurisprudence of selection patent has been more particularly evolved around pharmaceutical compounds. Section 3(d) was meant to prevent evergreening patents in India but it placed no bar on the general grant of selection patents. Even if there is a disclosure that is in principle novelty-destroying for each claimed feature individually, there can still be novelty for a combination of features that is not disclosed. This in fact is the basis of the "selection invention". Often, an unexpected advantage, that could not have been predicted from the prior art, is taken as evidence of an inventive step for "selection inventions" However, if the invention is found to be obvious, then an additional unexpected advantage can be at the best considered as a "bonus effect", which would not qualify to confer inventive step in selection. Determination of inventive step is normally subjective, which leaves a lot of scope for argument and different conclusions from the same facts. Therefore, an expert advice during the application process can be helpful in selection invention patenting. The fact being in the end, the applicant does get the benefit of the doubt, to some extent in such cases. Last but not the least the well-crafted original patent specification plays a decisive role in leaving room for selection patents. Expert advice in drafting the original patent application for such cases does result in carving of further selection patents. The applicant seeking a selection patent must demonstrate the inventive step by way of "inventive selection" from a field that is in general terms, already known. Lastly in selection patent the inventive step would lie in making the discovery that, what has been selected provides a genuine advantage over the generality from which it was selected and the advantage so found is one that could not be predicted.

The content of this article is intended to provide a general guide to the subject matter. Specialist advice should be sought about your specific circumstances.