Aleksandra: [00:02:14] Welcome everyone! We are live so when you are live with me, drop a comment that you’re here and let me know where you are dialing in from! That’s super important for me to know if I’m actually talking to people who are there or if I’m talking to people who are watching the recording. So, either way, let me know that you are there.
So today we’re gonna be talking about what FDA is saying about the use of whole life images for toxicological , pathology. So I have my here guidance document here with me. Let me know if you see this document. And basically I’m gonna tell you what, so this is guidance for the industry. Highlight guidance for the industry.
So what’s happening here is we have a document. They’re saying kind of what is recommended [00:03:14] it’s a draft guidance as well because there was a comment period where people were able to post comments. But they’re saying it’s just a guidance you know, don’t, like take it too literally.
The guidance is structured in a very nice way. I’m gonna show you but who am I even to talk about this? This is some guidance I am qualified, or I like self qualified myself to talk about it because I read this document like five times or probably maybe seven times. Analyzing this for different things that I do for my work, and I have also presented it internally at several occasions, so I decided to present it to you as well! So that you don’t have to go and read it.
Cause honestly, it’s not the most thrilling document I mean it’s FDA document all due respect the goal of this document is not to be entertaining. So I hope I’m gonna be a little bit more entertaining than just reading this [00:04:14] document. So, guidance for the industry.
Our guidance has an introduction, background, and questions and answers. What we’re gonna focus on are our questions and answers. Before I talk about those questions and answers let me know were you familiar for this?
Were you familiar with the fact that FDA actually said something about, oh, this is how we recommend you use digital pathology for non-clinical pathology for toxicologic pathology. Toxicology, pathology was everything that happens. The drugs before the compounds, are actually administered to humans.
So this is not the classical primary diagnosis thing where we have medical devices, it is more GLP where it is more, it is GLP Good Laboratory Practice regulations. This is Code of Federal Regulations. But let me know, did you know yes or no, that this is [00:05:14] out there and you know you can download it at any time?
Question one, what is whole slide imaging? whole slide imaging. This question probably we can do it very fast because everybody interested in digital pathology knows what whole slide imaging is. This is this, digitized digitization of glass slides. So here, generating two dimensional digital image. And here is an interesting part here two dimensional, we are not talking about anything three-dimensional, this is two dimensional.
So, you know, with just a flat image, a glass of histology slide used for routine assessment in generation of pathology report. So, the data In the non-clinical world in the preclinical talks path world whichever word you wanna use is the pathology report, this is considered data as per GLP regulations.
So if we are [00:06:14] using whole slide images for this particular purpose, we are, this is what we are generating our data off of. So yeah, and we can see here that due to some limitations of the scanning, FDA does not consider the resulting digital image to be an exact copy of the glass slide. It’s not an exact copy but it is there basically telling us here it is good enough. There is a term called faithful replica, that it is a faithful replica. It’s good enough to generate data on to generate pathology report.
And then the question comes, should whole slide images be retained? And a quick answer, yes, it should be retained. So what’s happening here is, if whole slide images in the previous question we saw, okay. Whole slide images are good enough to do your job as [00:07:14] a pathologist on if you’re a toxicologic pathologist, if you’re doing toxicological evaluation, good one green light.
Second question, should you retain them? well, if you’re doing your work on glass slides, then you retain them. If you are now doing your work on digital slides, logically you retain them as well. So GLP, if, let me know if you’re working in a GLP environment. Put a GLP if you are, and if you’re not, put an NGLP non GLP. Because for those who are in GLP that’s gonna be obvious.
But I want everybody to you know, take it home. So for GLP, there’s a lot the main focus or like the focus is on traceability of your data of on being able to retrieve and trace the information and retrieve why you got to the results. So in the [00:08:14] previous and yes, I will post the link to the guidance. I’m gonna post it in the description of this broadcast. So, totally it’s gonna be there. So, but long story short, you retained glass.
Now you’re doing your job on digital images. Yes, you retain digital images. And what do they tell about this? They say that if whole slide images are assessed in lieu of original glass slides. So if you are now doing your job on digital, then Yes, you have to retain it. So here is a little caveat because it’s during histopathology assessment and or pathology, peer review.
So currently what is peer review? Peer review is a review, like I’m reading a toxicologic study and there is a peer review as part of this whole study. Peer review is that somebody else is looking, if I did my [00:09:14] job well, basically a second pair of pathology eyes. And this is a part of the study evaluation.
It’s a process that’s happening routinely. It’s not that, let’s say, oh, I only have so many years of experience. There’s always. this Very often, not always, but very often included in the study design that there is gonna be a peer review. Somebody else usually from the organization that is commissioning this study is gonna check, okay, do they agree with my diagnosis?
They usually have more information about the compound I’m evaluating about previous studies that has been with the compound so they wanna see, okay, am I seeing, stuff that’s similar, how does it fit into the full picture? so they’re being my peer reviewer, but I am, there are questions.
And so I’m gonna finish this sentence and go to retask questions, so I am, but I still am the author of the primary report. [00:10:14] I’m the author of the data. This peer reviewer, peer review pathologist is giving me like advice, giving me feedback and then we have to agree on this, but the final decision or disagree, we can disagree as well.
If we disagree too bad, we have to figure, figure stuff out. But, the final decision is mine and that means that the. What peer review pathologist is doing and currently you can read a study on glass and have a peer reviewer view it, review it digitally. So their information is not really data because data is my report. Only
what I put in this report is data. So there is a little bit of controversy. Oh, do we really need to retain those slides if it’s only for peer review? There is a little bit of discussion in the industry because we don’t currently don’t do it. But this document is open for [00:11:14] discussion or for comments.
I think maybe this period is already closed, but everybody who wanted to comment it on that already had the chance to do that. And hopefully there is gonna be a different version later. So, yeah. another thing to this question too, and we have like eight questions. I’m spending a lot of time, a lot of time on my question, but, this basically disappears.
If your primary review is digital, then you have to retain those slides anyway and it doesn’t matter if peer review was done on digital, on glass, you retain your slides. So basically assuming at some point everybody’s gonna go digital and the digital evaluation, you don’t have the problem anymore.
You have to archive your slides anyway. So, let’s address a question from Rita for how long should we retain the digital image? Same as the glass slide. Yes, so this is all in the institution standard operating procedures, [00:12:14] but basically because this is what the data was generated from, it has to be retained as long as all the other study material.
So yes, it basically becomes another piece of data in this systems because it’s digital. But another piece of something that we have to keep with all the other stuff that we’re archiving for a study. And let us move to question number three. Question number three, if the whole slide images files out are retain, should the glass lights also be retained?
The answer here is yes. They quickly answered them the first sentence. So, and that kind of goes back to your question, Rita. The glass lights are study specimen and must be retained as study specimens after study fi finalization in accordance to this 21 CFR per 58. This is code of federal regulation that we are following if we’re doing [00:13:14] GLP.
And then Question four, what should be retained with respect to the whole slide image file? Should modified whole slide image files be retained? First of all, when you’re capturing those images we should not modify the file.
So then if you don’t modify the file, then there is nothing additional that you are retaining. But often modification happens when you are doing image analysis, and image analysis is being done for GLP studies as well. And then in that instance, you do retain the modified files. You do retain the or originals.
You do retain the modified, the purpose is slightly different. It’s not just primary evaluation that you use digital instead of glass, but so you have different type of outputs you’re maybe counting something or trying to quantify something with image analysis tools, and you then do retain it. But let’s check verbatim what they’re telling us.
[00:14:14] Yes. So, here the whole slide image files referred to here as the original whole slide image files should be retained. So the originals, we retain the originals, but any technical image processing modifications made to whole slide images files prior to being provided to pathologists should be documented and retained.
This is not really happening and you should get the original, as a pathologist, you should get the original, and so then you don’t have to document anything. If there is some smoothing, color manipulation happening before you get the slides, then of course you have to document it and retain this information.
But Here, they also say that viewing software should not allow the original whole slide image to be changed. So you are just viewing this slide. What you can do there, like you can do on your computer screen, you can adjust the brightness, adjust some contrast, but this is something you do under [00:15:14] microscope as well.
So this is allowed, not like regulated, doesn’t have to be documented because you are not changing the original image. So this is totally fine, here brightness contrast do not need to be documented or retained. It’s basically my personal preference as a pathologist I want it brighter, less bright, whatever. I can change it and not tell anybody about it ain’t that cool? I think it’s cool.
So now, here another question of ours. Should written procedure for whole slide imaging process be in place? Yes. Basically, if you work in GLP, you need written procedures for anything.
And the more I work in GLP the more I realize it, before when I started, I didn’t know how strict that is, but in GLP there’s even format mandated for how you sign stuff, how you put in the date, how you [00:16:14] correct input data. Like you have written procedures for everything. So obviously this is a big deal that you can now, look at slides under instead of under the microscope on the monitor.
So yeah, there should be written procedure for that. So, yes, that’s what FDA says. FDA says yes, and should be in place, written procedures. They might include scanning, validation, training, maintenance, software version control, backup, disaster recovery everything, everything, everything. But this everything, do not be terrified.
This everything is what you usually do for any computerized system and for anything in GLP good laboratory practices, you have to protect from all the stuff from virus protection, even, and disaster protection, anything. So, yes, the answer is yes. Question number six, should the whole slide image system be [00:17:14] validated?
I’m actually surprised that there’s no, yes here, but there is a qualifier. So as again, if we are using whole slide images, Instead of glass, which is in lieu of the original glass slides during histopathology assessment, and here, this peer review as well. Well peer review. Yeah, that’s fine. The only, the archival part was a little bit controversial, performed under GLP.
Then yes, we have to validate. Is that a surprise? No! because we have to validate everything for GLP, if you are doing non GLP, then you don’t have to validate, but for GLP, yes, any computerized system. So why am I saying a computerized system here? Because digital pathology system is a computerized system.
So basically this here, this guidance that we’re going through, Is kind of telling us how to deal with whole slide images as a [00:18:14] computerized system and how to apply whole slide images to what is already written in the 21 CFR part 58, there is a GLP regulation for computerized system and there’s exactly written, what do you do with a computerized system? So you take this digital pathology system, you say, oh, it’s a computerized system. What are the component of the computerized system? .
It’s this, that scanner viewer, whatever, uploading of slides, and you validate it. So, yeah, no surprise there. So going to our next, yeah. FDA themselves says, should be validated and maintain and matters specific to the intended use. That’s important as well intended use. Intended use, which is what you specify in your SOPs, and basically it’s specified.
A system should not be used in a way that’s not intended to be used. So important thing, but basically, yes.
And question number seven, how should [00:19:14] whole slide image files be protected, including when transmitted to external users? Well, so these are digital files. We have to protect them somehow.
So, If they are again, assessed in lieu of the original glass slides, because also at the beginning of this document it’s written there that it does not apply to pathology consultations, and education, and like all those ways that we already are using whole slide images for everything else, then primary study evaluation.
So for that we can use them no problem, but here if we wanna use them again in year of original glass slides, we have some stuff, to take care of. So, generation of backup file, chain of custody access controls, securing data systems and data transmission. Written [00:20:14] procedures, again in compliance with electronic records.
So this is then, this is becoming an electronic record. And in any, if at any point of time you have a question, give me the question, I’m happy to answer the question hopefully, or refer you to whoever knows the answer to this question. So but basically this slide becomes then electronic record.
So digital pathology system is a computerized system. When in doubt you go to, oh, what do I do with a computerized system? digital pathology slide is an electronic record. So when in doubt, How do you deal with electronic records? This is basically what you do with electronic records. You have a backup file, you like, have a chain of custody.
Not everybody can access and everything that you have to do to be compliant. So that is taking us to our last question and the [00:21:14] last question. Should the signed pathology report slash peer review statement, state that whole slide images were evaluated in a layer of glass slides? Without looking at the answer.
What do you think the answer is? Drop me in the comments. Just kidding. You don’t have to drop that in the comment because the answer of course is yes. Yes. Everything has to be stated. Everything has to be clear. Everything has to be traceable. So yes, we have to say this particular study was evaluated on digital slides.
If we’re doing GLP, if we’re doing primary evaluation for GLP on digital slides so again, yes. So we have some yeses. Good. Fantastic. Yeah. Also if pathology review, a pathology peer review is conducted on digital slides, then we say that it was done on digital slides. You know, given that we have validated everything, systems work, everything, [00:22:14] their SOP, standard operating procedures in place, and all that jazz, So this is what FDA is telling us about using of whole slide images.
And why I’m so happy about this document as well is because the, I mean here they are saying. Do what you should do for computerized systems. Do what you should do for electronic records. We kind of had that information already before because it’s not the first computerized system and it’s not the first electronic record that is being retained or dealt with under GLP.
But now FDA is actually telling our community, our digital pathology community, hey guys, here it’s guidance. It’s just a you know, draft guidance, but we know you are using it where this is how we are fine with using it. This is the [00:23:14] way you can apply your current regulation to this new form of data and new form of computerized systems.
So thank you very much FDA for putting out that guidance. Even though it’s draft, even though we are still waiting for comments there is a guidance for the industry. So for anybody who is questioning or who is asking themselves, can we use those whole slide images, like how do we use them? It’s written right here!
