One of the unique features of US patent practice is the existence of continuation applications. Back when US patents issued for 17 years from the grant date (and were not published prior to grant, a point to which I’ll return below), this meant that one could have a series continuation applications tracing their lineage back to a first application, so that one could have chain of patents with common ancestry but different expiration dates. (It was in this context that obviousness-type double-patenting and terminal disclaimer practice developed in the USA, but that’s not the topic for this post.) Patents earlier in the continuation chain couldn’t be cited against applications later in the chain.**
US practice also has something called a continuation-in-part application (CIP). By definition, a CIP contains new matter vis-à-vis the parent application. To the extent that the claims of the CIP are based solely on material in the parent case(s), the CIP benefits from the filing date(s) of the parent case(s). New material, and claims based thereupon, are only entitled to the CIP filing date, and in principle an ancestor patent could be cited against a subsequent CIP if that ancestor patent was published more than a year before the CIP filing date.
Under the old 17-years-from-filing regime, in which patent applications weren’t published to prior to grant, one could prosecute an application to allowance, but then rather than pay the issue fee, file a CIP with additional material. Provided there were no relevant third party intervening publications between the parent case filing date and the CIP filing date (and no relevant applicant publications from more than a year before the CIP filing date), this enabled applicants to continue to develop their inventions, and to supplement and expand their US patent applications, until such time as they were ready to let the patent issue, at which point they would prosecute the case to allowance and pay the issue fee. A further CIP could then be filed prior to issuance, or the applicant could from that point continue with a series of continuation applications.
In June 1995, the USA switched to a 20-years-from-earliest-filing regime. While this didn’t change the no-citing-the-parent-against-the-child practice in continuations (and for old material in CIPs), it did mean that one had to take into account the loss of patent term that results from filing a series of continuation applications. It also meant that since that time, when filing a CIP, one needs to consider whether or not the new material is patentable on its own, even in the face of the parent case. If the new material is patentable on its own, then it’s probably wise to simply file a new application, rather than file a CIP, because filing it as a CIP means you’re foregoing patent term – instead of 20 years from the filing date of the CIP, you get 20 years from the earliest non-provisional application in the chain.
I actually encountered a case like this shortly after I got my USPTO license: an applicant had filed a series of continuations, and then, after June 1995, filed a CIP and got a patent that claimed the new material – in this case, some chemical compounds that showed promise for treating a particular disease. But because of the CIP status, and because the earliest application in the chain had been filed five years earlier, the patent would expire after 15 years from the CIP filing date, not 20, even though the new material was patentable on its own at the time of filing. As a result, big pharma companies didn’t want to invest. It wasn’t my mistake; I was merely given the honor of telling the client it was screwed.
In March 2001 CIP practice took another hit. That’s when the USPTO started publishing US applications prior to issuance, nominally 18 months after filing. What this meant was that if you filed a CIP more than one year after the parent application published, that parent application would be available as a printed publication against any claims in the CIP that were based in the new material in the CIP. While the publication of one’s own earlier application was already a concern for applicants who filed outside the USA (for example who filed PCT applications that published 18 months after the priority date), from March 2001 it was something with which US-only filers also needed to concern themselves.
Since this is all old news, why am I writing about it now? Because, remarkably, this is still an issue that some people get wrong, as I saw this week.
A colleague here in Israel for whom I have in the past filed PCT national phase applications in the USA contacted me. His client, an Israel-based company, has a US subsidiary, and that subsidiary has an in-house US patent attorney. That attorney had suggested filing some material in a CIP of a pending US national phase application, and my colleague asked what I thought. I told him I saw no advantage to be gained. If the new stuff was not patentable in view of the parent PCT, which published several years ago, then calling the application with the new stuff a CIP wouldn’t help, because that PCT was citable against the new stuff. And if the new stuff was patentable notwithstanding the published PCT, then why foreshorten patent term by making the new application a CIP of the US national phase application? File a standalone and get the full 20 years!
I remain amazed that there are practitioners out there giving their clients bad advice.
To be fair, there are situations in which it makes sense to file a CIP, but those situations are quite limited. The most common such situation is when there’s new material and you’re still within a year of the publication of the PCT application. In that case, you can incorporate the new stuff, and then file your application in the USA as a CIP of the PCT application, what some of us refer to as a “CIP bypass” (a term that I think was coined by Sam Helfgott, who retired from practice ten years ago or more). By getting a filing date within one year of the PCT publication, that publication is not citable against the CIP; by making the new application a CIP of the PCT, the new application gets the benefit of the PCT date for any material that was already in the PCT and that might be claimed.
**Many other jurisdictions allow divisional applications, and in some of those jurisdictions, such as Israel, it’s possible to file a series of divisional applications in much the same way that continuations are used in the USA. However, there’s also a thing called a divisional application in the USA, and in particular circumstances calling an application a divisional rather than calling it a continuation application can be advantageous for purposes of obviousness-type double-patenting. See 35 U.S.C. §121 and e.g. Pfizer et al. v Teva, 518 F.3d 1353 (2008) (Fed. Cir. 2008).