Anders - Just my opinion but here it goes (sorry for how long this is). The ever lengthening amount of time since data lock with no TLD announcement increasingly makes me think they are planning to announce TLD via a simultaneous peer-reviewed publication. This is not the "normal" way of doing TLD announcements, but this is not a normal dataset. I think the multiple analyses planned (potentially getting at the same issue using multiple statistical approaches), the multiple endpoints, the use of historical controls which needs to be fully explained, the possible failure of original endpoints despite apparent trial success compared to other similar trials, etc. all may require more detailed context than is possible in the typical standard 2-3 paragraph TLD press release. This context will be critical to making sure that the results come across positively rather than incorrectly making the trial look like it failed or is questionable, so it is definitely worth the wait to do things right (as DI has also apparently stated). If the statisticians have been moving cautiously to make sure that all of the analyses are complete and correct, I think they should have had all results, tables, and figures needed to begin writing a publication available by approximately mid-November (possibly sooner if they have been dedicated solely to working on this one project). We know from LL's recent talk that they do in fact already have the survival curves (individually and pooled) necessary to do what I suspect might be the most time consuming part of the primary analysis - comparison to external controls. It may may have taken them some time to identify and agree on the right comparison trials and construct these survival curves (I'm not sure sure whether they had easy access to the raw patient-level trial data from other trials). So, I assume they had the complete results together by about mid-November. I further assume that LL will be first author (as in the blinded/blended paper) and that she would be the one drafting the paper, possibly with the primary statistician simultaneously working on the statistical methods and results sections. LL always seems to be quite busy, but given that she knows this literature quite well without having to do an extensive literature search, I can see her finishing an initial draft in as short as 2-3 weeks (beginning of December). Next, she would have to send the draft out to 68 or so co-authors for comments and edits. If I were in her shoes, I would give the co-authors a two week deadline (which is a realistic minimum for busy physicians) to send back comments. Every time I have written a paper with a lot of authors, it has been like herding cats, and there are always a couple of people who don't respond until they have been reminded multiple times. LL will be limited in her ability to finalize and submit the paper until every author responds, because all authors will be required at the time of submission to click a link saying they did indeed contribute to and agree with what is presented in the submitted paper and they all will also have to fill out online copyright assignment and conflict of interest forms. The article will not be sent out for review until all authors do this. Someone will always be slow to take care of these requirements, and with 60+ co-authors, I can see this taking up to a couple of weeks between submission and actually being sent out for review. So my guess is that they by today would have gotten to the point of having an article submitted. As I have said in prior posts, if their original planned PFS and OS outcomes (DCVAX vs. the internal placebo group) are not significant and the only significant findings are the comparisons to historical controls, getting this accepted by NEJM or Lancet may be challenging. However, I assume they would want to try one or both of these top journals first. These high end journals do an internal editorial board review of submissions before the paper is sent out for peer-review. If the editorial board thinks the paper is unlikely to get accepted eventually, they will typically within a week or so of submission contact the corresponding author and reject the paper before it is even sent out for review. I think LL would at least try to send this paper in to one or both of these top journals. If they get editorially triaged at a top journal, they would then immediately submit the paper unchanged to another journal. If NEJM or Lancet did send it out for full review, I think it would be on a rapid timeline, where the authors would get reviews back within 2 weeks. No paper is ever accepted without revisions, so there would be a process of another approximately 2 weeks (if they are trying to rush it) to revise the paper and get everyone to ok the changes, followed by another 2 week review process (and hopefully acceptance). The best case scenario assuming they submitted the article by 12/15 would be formal acceptance by one of these two top journals by mid-January (although holidays often will slow this process down). Once the company has an acceptance letter in hand, they could in theory announce TLD in a PR saying the paper has been accepted by XX journal, and provide some additional context in the PR based on what is written in the article. My guess however is that they would want to wait to release TLD until the manuscript actually has come out in electronic preprint form (which could be as short as 3-4 weeks after formal acceptance based on my experience). That is the best case scenario at a top journal. If the first submission is not successful, each time the article is rejected we are probably looking at adding a minimum of another 4 weeks to the timeline above. My guess is that this paper will be accepted by some reasonable quality journal by around the end of January, because the trial is unique in how long it ran (5 year survival is for all practical purposes actual survival rather than K-M estimates), how deadly the disease is, and because it suggests a rare success of a vaccine approach to cancer (so it is quite novel). So, there is a lot of guesswork in this, but considering all information available, I think the publication could come out by mid--January at the absolute earliest, and possibly as late as end of February if they end up on a torturous path to acceptance. This is all informed guesswork. It is always possible that if the publication process looks like it will not be smooth, they could change plans and release TLD in a PR earlier. If we have not heard anything about TLD by the first week in Janury, I think it will be pretty certain that they are waiting on a journal acceptance to release TLD.
