This is our discussion with Arvilla Trag which was recorded for the Solutions OP Clinical Trials podcast. You can listen to the original recording here. Please note that the interview below is adapted and not a transcript. This was done to improve readability.
Interviewer: Welcome to today’s episode of the podcast. In regulatory affairs, we always talk about what you should not do because you end up in trouble, and this has created a bad reputation for the whole regulatory affairs field, that we’re some kind of boring killjoys who constantly nag you to do things because of some regulation.
The fact is that many of us are very funny, but we haven’t had the opportunity to show that side. And this is why today I have invited a very special guest, Arvilla Trag, an expert in Chemistry, Manufacturing and Controls (CMC) with over 30 years’ experience in helping pharma companies get out of the FDA CMC hell.
However, instead of telling you what not to do, today we’ll tell you what you should do if you really want to piss off your FDA reviewer. If you follow our instructions, we can guarantee your application will fail with the FDA. In order to do that, Arvilla will put on his Dante Alighieri hat, and she’ll walk us through the nice circles of the FDA CMC hell.
Before we start, I’ll ask her a few more questions to tell me about her experience, what she does, and how many people she has rescued from the FDA hell so far.
Arvilla Trag: I have been doing this for 34 years next month. I was a research lab technician, and my salary was funded by a grant. The grant wasn’t renewed, and I heard about a job in regulatory affairs and quality assurance at a very small vaccine R&D facility. I didn’t know what regulatory affairs and quality were, but it involved working with a lab, so I said, “OK, maybe I could do it”. I applied for the job and got it because I didn’t know anything and they could pay me less.
I was there for about 6 months, and I realised, “I love what I’m doing here”. I have stayed in regulatory and compliance ever since. It has just been, if you’re going to work, you might as well enjoy what you’re doing, because one of the circles of regular hell is getting up in the morning and going to work, oh my God, I don’t want to go there.
I love writing CMC sections for BLAs. I love writing briefing books for FDA meetings. I love doing that kind of work. I’ve done over 275 compliance audits in my life, and I’ve done enough of them. I’ve run into some people who’ve said indescribably bizarre things to me, “As long as the process keeps working, it’ll be fine”.
I’ve got some funny stories and some just plain scary stories, because these people are out there. I’ll be happy to answer all these questions.
Interviewer: Thank you so much, Arvilla. You had too many audits. I have to say that’s quite too much by everyone’s standards. Let’s go to the different circles of hell and see what’s going on in purgatory, what’s happening in each one of them. If you want to, just walk us through, give us some examples and whatever you would like to add on that.
Arvilla Trag: Well, purgatory is kind of a clearing house for people on their way further down. And right now it’s in conflict because it is so crowded. There are some who want to turn it into a great big housing development. Of course, there’s always somebody who wants to develop housing, no matter where you are.
Those who opposed the housing development insist it would create only a 10th circle of hell, and they’ve got quite enough. Then there’s a third faction that would like to see a 10th circle of hell but cannot agree on where it should be placed among the other circles.
The logistics of building a housing development or a 10th circle of hell in hell are challenging too, because most of the contractors in hell are down in the 8th circle, where the fraud goes. The paperwork to get a temporary transfer to another circle to do any construction takes an eternity. And it’s been over 2000 years since there was a carpenter in Purgatory. That’s what’s going on in Purgatory.
Now, my first client had me navigating the 7th circle of hell, which is failure to follow FDA advice. He was extracting a specific protein from elective reduction mammoplasty in a warehouse he had rented, with no aseptic controls whatsoever, and sending it to stage 4 breast cancer patients to store in their refrigerators next to the Grey Poupon and self-inject intravenously at home. All of this was being done without the benefit of an IND, a clinical protocol, monitoring, or physician supervision.
Long story short, a very infamous female politician with the initials HC interceded on my client’s behalf when the FDA was about to send him to prison. He was not prosecuted but had to comply with very strict conditions set by the FDA. One of those was that he could not enrol any more patients. And a monthly report on the status of the patients had to be submitted to the FDA, but he couldn’t do it. They didn’t trust him, so they said he had to get a consultant. So, I was the consultant.
At the time, he had 24 women injecting themselves with this stuff, which was going out with like 9.5 EU/mL of endotoxin and such. I quit working for him when I discovered that he was not reporting patient deaths to me. He had been enrolling people without telling me, and I was sending reports to the FDA saying everything was like this when it was actually like that. And I told him, “You’re not paying me enough to go to prison for you”. I wasn’t particularly interested in being in the 7th or 8th circle. So, he’s going to end up in either the 7th or 8th circle of hell, but right now he’s a regulatory consultant in Northern Michigan.
Although I wouldn’t want to live there, the eighth circle fascinates me because of the fraud. When the FDA negotiated subletting portions of each circle to be specific for their punishment, you should see that contract, and they did this in 2021. They also created an area in each circle where you can observe what is going on in the rest of the neighbourhood.
The 8th Circle was originally intended for fraud, specifically malicious deception. This circle is divided into 10 ditches or bulges for pimps, flatterers, simoniacs (people who sell or trade ecclesiastical favours), corrupt politicians, hypocrites, thieves, evil counsellors, sowers of discord, and falsifiers, basically Washington DC.
Hell is filling up so fast these days that some of the most egregious sinners already have a reserved spot, even though they’re not dead yet. They’re in a designated circle with a sign that says “reserved for,” and they’re eventually going to end up in that spot, kind of like reserving a table at a restaurant except not so much fun.
Reading these reserve signs is very informative. Now, I can’t mention any specific names due to my three-way CDA agreement with the FDA and hell, but I can tell you that the current manager of FDA Hell is Joseph Mengele. And the next one is going to have the initials AF. It was very prominent during the pandemic, or if one chose to believe that. Now, the best way to piss off the FDA is a tough one because there are so many ways to do it.
It’s hard to choose which is the best way. I think it would partly depend on your product and your history with the FDA. If you have a history of, let’s say, making claims that are not necessarily accurate, such as telling the FDA in March that you would file your BLA by September, when the reality was that, as you spoke, a tech transfer was underway, moving from one CMO to another, and there wasn’t a snowball’s chance in hell, you should forgive the expression, of actually getting to Process Performance Qualification (PPQ) in 6 months.
When the FDA is told that you’re going to file at a certain time, they reserve resources so that it can be done in a timely manner. So, 2.5 years later, when this same client wrote to the FDA and asked for a pre-BLA meeting, the FDA responded, and this proves that the FDA has a sense of humour. The FDA responded by saying that “In March 2017, you told us you were going to file your BLA in the fourth quarter of that year. Considering the amount of time since then, we can only assume that your PPQ batches have matured”. And I don’t think I’ve ever laughed so hard about anything I’ve read from the FDA.
Then there’s the trifecta, where if you can do 3 things for that merit, they have to be 3 different circles, not 3 things for the same circle. It’s got to be 3 things for 3 different circles. You hit the trifecta, and that pisses off the FDA. So, for instance, bloviating, dead-end or loop hyperlinks, and poor graphics would combine circles 1,3, and 4. There’s a trifecta, but if you just don’t want to bother going to the effort of getting all three circles, or even three circles in.
The absolute best way to piss off the FDA is to go to them, ask for advice, listen to it, then ignore it and go back to tell them it didn’t work. That pisses them off because the FDA does not give advice lightly. When they do, they expect you to listen to it. And some people do, and some people don’t.
Interviewer: Yeah, they just try to see what’s happening if they don’t listen. I think sometimes people just want to be risk-takers, they want to do it themselves and see what happens.
Arvilla Trag: A lot of people genuinely don’t believe in the power the FDA has over drug developers. You know, you want a meeting with us? We’ve got an opening in 3 years. Actually, they’re required to grant meetings within a certain window of time after you request one.
But, you know, there are certain people you should never piss off, like your hairdresser, your dentist, your gynaecologist, and your mother-in-law. But one of the groups you shouldn’t piss off is the FDA.
It’s just that it’s a win-lose situation. They’ll win, you’ll lose. It’s their ball, their rules. What else can I tell you? What else would you like to ask me?
Interviewer: Tell me a bit more about the writing style. How would you prefer people to write the section and present all the information?
Arvilla Trag: Oh my God, I just did an 8-hour workshop on that. Oh yeah, and at my ripe old age, staying awake for 8 consecutive hours was no small trick.
Write for your audience. Don’t write like you’re writing a journal article. Don’t write like you’re writing a children’s bedtime story. I’ve seen people draft documents in both ways. Write for your audience, with the understanding that the FDA does not need certain things explained to them, and that they do need other things explained.
When you put together a glossary and list all your acronyms, do not include FDA, USA, CMC, and similar terms, because FDA knows what FDA stands for. When you are writing, make sure that you’re not making any claims. Make sure you’re not bloviating, and write in chronological order. Don’t say something like, “Before we did this, we did that”. Now, just write at the end, first we did this, second we did that, third we did this. Do it in sequence.
And when you are putting a submission together, make sure there are no dead ends or loops in your hyperlinks, and that they actually go to where you say they’re going to go. And do not attach the guidance documents because they have them. I have seen people cite the guidance document, so they were going to attach it just so FDA would know, and you really don’t need to do that now.
Interviewer: How about the resources? Do you have any recommendations for writing?
Arvilla Trag: Well, aside from the fact that everyone should hire me because I am really good at what I do, it beats the hell out of doing audits. You need somebody who knows module 3 inside out.
Unfortunately, ICH is revising it, so it’s coming out more like Frankenstein’s module than anything else. I’m serious. They discuss the working cell bank long before they even get to the master cell bank. Talk about not being in chronological order. They describe analytical methods in one section and, way over in another, they’ve got a validation of those methods. Nothing in the control section is together anymore. The specifications, justification, and so on are spread throughout.
And this is because you cannot take a 26-person committee and put them in a room with 100 typewriters and get the same good results you get with 100 chimpanzees. Everybody had to put their two cents in on this revision of M4Q, and it’s a train wreck. All they had to do was add some sections to the existing module, and it would have been much clearer and made sense. There is no logic to the M4Q revision. It’s only logical if you are someone on a committee who has never actually prepared a module 3. It sounds like we should put this here and put that there. So, you know, these kinds of experiments are grouped together.
No. You do not put your analytical validation separate from your analytical methods and the justification of specifications. Hyperlinks are great. But people should not have to bounce back and forth all over a freaking document that can be hundreds of pages, just to find out what’s going on with an analytical method or how you got your working cell bank into being to begin with, because you can’t describe the master cell bank before you do the working cell bank in this format.
Interviewer: And tell me why it’s in the 9th circle that missing a required section is so special?
Arvilla Trag: The FDA electronic gateway for submissions is set up to scan for the different sections. When you submit your BLA, you’re going to submit a non-clinical section, a clinical section, a manufacturing section, attachments and appendices, and then your module two sections and module one. It’s looking for this sequence of numbers, and they all have to be there. Otherwise, it’s going to reject it because you haven’t submitted a complete submission.
Should you move things around or leave one out, let’s say skip 3.6 and go directly from 5 to 7. Basically, what happens is computer apoplexy. And the FDA gets really pissed because you’ve screwed up the system, and now they’ve got to get their IT people in there. Now, I don’t know if that’s actually true about bringing in the IT people.
They’ve put together module 3 for a reason, the way they’ve structured it now, considering that before the ICH module 3 guidance, the only guidance that people in the United States had for writing a marketing application or an IND was 7 paragraphs in 21 CFR 314, 7 paragraphs to tell you how to write all this.
So now, I think the current guidance on module 3 is 24 or 26 pages, and the new one is almost double that. But they’ve included so many sections that say, if applicable, optional. I mean, you could just put a list of things that are, if applicable and optional, separate from that, so it’s a little bit easier to read through, but the people in that committee are definitely going to one of the circles of hell.
I haven’t quite figured out where they’re doing for sure. Bloviating is always popular. Probably, definitely, the 8th circle. Renumbering and reordering sections of the CTD. Yeah, that’s where they’re doing.
Interviewer: Oh yeah, that’s a good one. So, which is your favourite circle of hell, or maybe you have more than one that you prefer?
Arvilla Trag: I kind of like the 8th circle because it’s interesting to get on the, and they’ve got a viewing deck. You can sit up there and look at what’s going on and who’s getting what. Believe me, the 8th circle is no place to go. It’s pretty uncomfortable.
But the 9th circle, the frozen lake, where that is for omission of a required section, you can be frozen in the lake up to your neck, or you can be frozen in the leg up to your ankles, except your ankles are what’s sticking out. Or you can be completely under the ice. It depends on how many sections you left out. That’s pretty ugly.
And Purgatory is always interesting because it’s nice to hear all these people say, “Why didn’t I listen to my consultant?”
Interviewer: We will never know that. That’s the billion-dollar question. And what is the one thing people should do if they really want to piss off their FDA reviewers? What is your special recommendation for that?
Arvilla Trag: As a special recommendation, I was once given a process development report that was 300 pages long. In this report, there were 35 pages about the history of bioreactors and 42 pages about the history of impellers. When I went through it and highlighted what was pertinent to process development, there were 11 pages of actual information. So, submit that to the FDA as an example of your process development, then sit down and just wait for the grenade you shoved up your nether regions to go off.
I had a colleague, and we were working on this project. We would talk every afternoon after we had finished the regular work: “So, can you believe this? No. What did you have in your documents? Oh my God, you didn’t really? It was insane.” And this guy is as bad as me because we’re both in a race to screw up our retirements. We both want to retire, but we can’t seem to say “no” when somebody brings us a juicy project. And that is precisely why, instead of starting tomato seedlings in 2 weeks, I’m going to China.
Interviewer: Oh gosh. That sounds fun.
Arvilla Trag: Not so much. When you get to be my age, even flying business class is miserable. I get flat spots in places I didn’t use to, and it’s 16 hours to go from Dallas-Fort Worth to Shanghai. That’s a long time in an aeroplane. I bring my reading. I bring my knitting because leisure time for leisure reading is a humongous luxury for me. They wouldn’t allow knitting needles on an aeroplane until they realised that depriving a bunch of little old ladies of their knitting was more likely to cause plane crashes than a bunch of terrorists.
You know, one thing you never mess with is a knitter’s knitting. You just don’t. I mean, it’d be like messing with Bobby Flee’s knives. You just don’t, so. Where the hell was I going with this?
Interviewer: How you had to travel, you’re going to travel to China to have some more fun with a fun project that’s coming.
Arvilla Trag: Well, I’m 71 right now, so I just turned 71 a couple of days ago. I’m telling you, when I was in my 50s, even in my 60s, I could do this kind of travel and it didn’t affect me like this. Now I have to leave so that I’m on the ground for 2 days before I work, in order to reset my internal clock. And it runs on a 24-hour military time. The rest of the world uses the 12-hour system, and we’re all screwed up. Plus, China has a single time zone, even though the country is as big as it is. It’s humongous. They have one time zone from border to border, east to west. So, at 12 noon, the sun is just coming up on the western border of China.
I have no idea how, I mean, time, at least as measured by clocks, is an artificial construct. I usually just get up with the sun, except in the dead of winter, when I have to. It doesn’t get light here until about 8:00 or 8:30, so I have to be up before then. I’m usually up at 5 in the morning. But I don’t understand why they don’t have time zones, and I don’t understand why the airline I typically fly to China doesn’t have any proper padding in their business-class seats, because they are as hard as a damned board.
Interviewer: I hope at least they have their documents translated into English, so you have something to review, because otherwise you have to polish your Chinese.
Arvilla Trag: Well, there’s translation, and then there’s translation done correctly. The company’s policy is to have the SMEs who write the procedures do the translation. But they don’t bother to check whether the SMEs have good English or are desperately struggling to figure out how to say anything in English.
And I mean, their SMEs are very good people. They know their stuff, but not all of them speak English. And no matter what happens, their English will always be better than my Mandarin, because I don’t speak it. I can say “thank you”. That’s it. Because I’m a polite person. No matter what, I learned the words for “thank you”.
But you can’t necessarily rely on the translations. I was talking to a colleague last night who’s working with another Chinese company, and one of the translations was because they put the Mandarin, then the English, then the Mandarin, then the English. Well, one of the translations didn’t look quite right. So, the guy called a friend of his who worked at one of the company’s China offices and asked him to look at it.
The SOP was about human resources in Chinese, but in English it had been translated to be about a laboratory practice. Now, if the FDA sends in a Mandarin-fluent inspector to that facility and sees the discrepancy between the Mandarin and the English, there will be people being pickled and fried.
Interviewer: And they’ll question everything else because once they discover something, they’ll pick up on the whole application and everything else, as they expect you to provide them with adequate translation. And things like that are actually very common with Chinese companies that try to submit applications to the US and other English-speaking countries. If you don’t have a proper translation, that could be a real deal-breaker.
Arvilla Trag: This is why the FDA has required that if a document is originally in a foreign language, it be translated by a certified translator, rather than just having somebody in the lab who says, “I took 2 years of English in high school, I can do the translation”.
I took 5 years of French in high school, and I speak enough French to be very comfortable in France when asking for directions. I was even able to have a gift drop-shipped to my father from a little shop in Dijon. My father at the time was smoking pipes, and they had a pipe ashtray, which is very different from a regular ashtray. He’d been looking for one for a few years, so I was able to get it drop-shipped to him without saying a word of English, because the proprietor didn’t speak English. But I would never consider submitting to any regulatory agency in French because I don’t speak technical French.
Interviewer: Yeah, it’s completely different. It’s like a completely different language.
Arvilla Trag: It’s just like writing for journals and writing for regulatory submissions, which are like writing in two different languages. One of the things that drives me nuts about journals is that they will have paragraphs that take up a whole page, when they should be no more than 20% of the page.
If you keep writing a paragraph like that, nobody can keep track of where they are in the paragraph. And it breaks all the rules of proper English, grammar and style that you were taught in 8th grade. And it’s just, you’re not making life easy for your readers. Whereas I insist that things be grammatically correct as well as technically correct.
If you make it easy for your reviewer, the reviewer is happy. And a happy reviewer is a positive reviewer, and a positive reviewer is less likely to beat the living crap out of you for the mistakes you did make.
Interviewer: And I think that’s a very good final one. I’m just going to do a quick summary, and I’ll say that if you really want to piss off your reviewer, first make sure you miss the required sections. Write some word salad, so no one knows what you’re talking about. Include some history of manufacturing that kind of thing that will be very exciting for the FDA reviewers because they haven’t read them for a while.
Just don’t use any source materials or references. Don’t include any links or make sure your links go nowhere. I think that’s a brilliant one. You click on the link, and it goes to a blank page or to an adult website.
Make sure you include really bad graphs, maybe of your cats or dog, something exciting for them to see. If you have tables from a company that’s developed a product in China, make sure they’re in Chinese, because they can spend some time translating them. After all, they have plenty of free time. And don’t forget, if they give you scientific advice, just ignore it. You know better than them. Just ignore it. Why would you listen? You’re an independent person. I think that’s the best way to conclude our recommendations on how to fail your application.
