Patent Busting A Literature-based Approach
July 19th, 2025 By John Widen
I was told at multiple companies to never use the words ‘patent busting’ in a company email or presentations
even though it’s not illegal. Infringing on patents is of course illegal, but that is the point of ‘busting’ a patent.
Instead, we opt for the terms ‘literature-based approach’ or ‘information-driven approach’, which I’ll admit sound more
eloquent. At this point, I should probably give a disclaimer that I am NOT a lawyer and therefore cannot give legal advice.
However, I am a medicinal chemist! So, I’d like to share my experiences taking a literature-based approach to
begin a small molecule drug discovery program. Some folks in biotech and the pharmaceutical industry might look down
on this approach because scientists think patent busting strangles innovative science, is a waste of resources, and
is largely an attempt to ride on the coat tails of a drug that is already successful. There are many blog posts about
this topic already. This one for instance by Derek Lowe. Some of these arguments may have some truth to them but I
disagree with these statements generally.
BUT, I’m not here to discuss whether companies should be patent busting or not. I want to discuss HOW to patent bust.
The simple truth is that whether it’s a noble cause or not, it is fun! I love the challenge. I do think it is a worthwhile endeavor
because nothing is perfect (especially drugs) and there is always room for optimization. I’ll talk about how to search for relevant
patent applications to start a small molecule drug discovery program in this post and I’ll focus on some specific patent applications
and concepts in subsequent posts. It will be a blog series! Very exciting…SO LETS GET INTO IT!
Searching for chemical matter as potential starting points for a project can be daunting. There are two places to look.
The scientific literature and patent applications. For this post I will focus on searching the patent literature. The same principles
can be applied when searching for either. Patent applications aren’t peer reviewed. However, if someone went out of their way to write
and pay for a patent application submission, then it might be of interest for your project. Chemical matter in scientific literature always
makes me wonder if there isn’t an associated patent application. The answer is usually obvious after reading the paper and looking at the
chemical matter. But, I’ll let you decide what is an attractive starting point for your project!
Searching for relevant patent applications using search engines like Google Patents and SciFinder can be cumbersome.
Many of times have I opened SciFinder and typed in a drug target to find THOUSANDS of patent applications published. But thankfully, most
search tools provide easy methods to cut down the number of patent applications to look through. The very first thing I do is reduce the time frame.
Patents expire after approximately twenty years. There are nuances to this because companies can use tricks to extend patents but those are
generally rare and irrelevant. You are likely working on a newer target that does not have meaningful chemical matter that far back.
So, I won’t talk about that. I will count back twenty years and make a cutoff. Right now that is 2005. One more note is that I’ll shorten
this up even more at first to focus on most recent patent applications
(you can also just sort by date, easy peasy). Great, so you still likely have thousands of titles and abstracts to search through.
Most search engines have Boolean logic search terms. You can add multiple key terms or words separated by ‘AND’. You can also combine
terms with ‘OR’ to look for multiple synonymous terms at once. This is useful because if you’re looking for small molecules there is a good
chance that ‘small molecule’, ‘chemical structure’, ‘formula’, etc. are within the patent application. Additionally, you can narrow these
terms to only look in the title, abstract, or claims to narrow down the search further.
Another note is that if you are searching for a target that is, for instance, Stimulator of Interferon Genes (STING) you should wrap
the term in quotes to search for the entire term instead of each word individually as most search engines handle that query. You do not want
to look for all patents with the words stimulator, of, interferon, and genes anywhere in the patent. Sometimes you can get away with just putting
the acronym in for the target but it will always be defined. No need to look for things related to treating bee or wasp stings. Or a patent
regarding running a sting operation. Or taking the sting out of a sunburn. You get the point.
So far, we have searched for the terms “Stimulator of Interferon Genes” AND “small molecule”. That gives >5,000 hits. Way too many.
One other way to narrow down your list is to remove patents with certain terms. Typically, anything with the term ‘method’ or ‘methods of
treatment’ in the title can be removed because they are not going to focus on unique chemical matter. In Google Patents, using the “-method”
term looking only in the title is a good way to throw these non-useful patent applications out. The '-' (dash) is to remove documents containing this term.
And there are SO many of these.
This addition narrows it down to 27 patent applications! NICE! A great starting point. But wait…That seems a little fishy because
that is very few patent applications for a very popular target. You do have to be
careful because this will erroneously eliminate patent applications that do not have the term ‘method’ in the title. I have written a letter to
Google and awaiting a response. In the case of searching for STING chemical matter the “-method” removed relevant
patents from Boehringer Ingelheim for some reason. Be careful. Regardless, these filters will get you started and once you get a feel
of the landscape, going back and looking through other lists is not so daunting.
Another useful filter is narrowing down by company or institution. This term is either ‘institute’ or ‘assignee’ in patent literature.
If you know that there are certain players that have published patent applications, I would absolutely start there and expand.
The last filter I frequently use is not full proof but works well to narrow down the field to meaningful patent applications if
you are still left with thousands on the list. I only focus on patents in English. Worthwhile patent applications from another country
will have an English version because they want their chemical matter protected within English speaking countries. And for that you
generally have to nationalize a patent application. Meaning, there are versions of the patent in many different languages.
If there are other relevant applications you can always go back, but for now that helps get started.
After narrowing down to hundreds of patent applications that involve your target of interest, it’s time to scroll or download
a list to peruse. I ignore immediately any applications with the word ‘methods’ in it. As mentioned previously, it’s not always easy
to get rid of all ‘methods’ or ‘applications’ patents. Any title with these terms is a tipoff that the patent application is not focused
on novel small molecules. Many, many of the titles in the list will have this directly in the title allowing for easy triaging.
I will also focus on the abstracts and if there is no mention of the target in there, it’s not necessarily to be thrown out but
unlikely to be useful for your starting point.
Okay! So, you now have a folder of downloaded (mostly) relevant patent applications! That’s a great start.
Now it’s time to start looking through them to find the gaps in claims and identify starting points for your program! This is the fun
part in my humble opinion. However, this blog post is already running long. If you’ve made it this far, I commend you. Thanks for
sticking with me on this boring topic of searching patent applications. I just thought it would be useful information to cover
for thos that are unfamiliar with the process.
The next post in this blog series will focus on identifying gaps in patent claims and looking for good starting points.
I’ll discuss my approach and probably stick with STING as my example case.
If you have any other tips or tricks to searching for patent applications, drop me a line at jwiden@chemjam.com.
I’d love to hear about them and can include them in this post.
The site does not have a comments section yet! Hopefully, very soon! Until then please drop me a line at jwiden@chemjam.com.
If you provide comments on my articles I reserve the right to post them on this website as additional commentary. My goal is to have an open discussion!
