Aleks: The digital part of the pathology workflow, um, most of the time starts with the scanner. Scott, what were the, like, obstacles or lessons learned that you guys had to go through that you could then let the Proscia team know about? So that, was there anything like that?
Or everybody had to do their own thing because you were approaching it from different sides, uh, hardware versus software, although, uh, Hamamatsu also has the software component that was part of your submission.
Scott: Right.
Aleks: Any like, any hurdles, any problems, anything that you were able to say to let the colleagues at Proscia know that, Hey, pay attention to this or that, or anything like that?
Scott: Well, as, as you can probably understand, there are many hurdles, um, there are many points of concern that come up that you never even thought about. So, uh,
Aleks: you mentioned FDA. It’s kind of like equals. It’s greater or equal [00:10:00] hurdles.
Scott: Well, they’re different. Let me say that. So, for example, on the scanner side, we have to be remarkably concerned about what we call the pixel pathway.
The pixel pathway needs to be defined to the FDA, and then we need to maintain the pixel pathway. So, that involves what is identified within the scanner configuration, what the pixel pathway is. Once that is defined and then accepted, let’s say, we need to maintain that. So, when working with Proscia, naturally, we’ve known Proscia for quite, quite a long time.
And, um, when they came to us, For their side of this FDA process, we were very happy to help them. When we went through our tasks to, to do what we needed to do, we, as I said, defined the pixel pathway. When Proscia approached us, then we needed to maintain that pixel pathway when we were started to, to acquire data for Proscia.
So that’s [00:11:00] always, I don’t want to say difficult, but we need to maintain that and be assured that we’re maintaining that. And Proscianeeds to be assured that we’re maintaining that.
Aleks: So how do you, that is a, that is a super interesting thing. Um, specifically in the I want to call it new age or new era or more emphasis of interoperability.
The pixel pathway just like a super high level explanation is like from generating this pixel from the glass slide to actually seeing it by the pathologist. And it goes through a lot of transformations, a lot of different things, hardware components, software components. And now we’re introducing, so let’s say we had it defined, uh, in the first place.
Because a scanner gained, uh, the clearance, but now we’re introducing somebody else who also wants to be cleared, but wants to leverage this already existing pixel pathway. How did you guys do it on both ends? Are there like processes, mechanisms, [00:12:00] ways of. Enabling this kind of integration.
Amanda: Hamamatsu did provide us with an SDK.
So basically they gave us a developer kit for, uh, ingesting their images out of their pixel pipeline. And then on the Proscia side, so essentially in that cleared system that the FDA sees as a closed system, we would be replacing the Hamamatsu, um, MCU view. in the viewer, right? So the Proscia viewer is replacing the not close system.
Aleks: Okay, so you basically cut off the pathway. It stays as it was on the Hamamatsu sign, and Scott, correct me if I’m wrong, but I’m just like trying to figure it out for somebody who has not been in the trend yet, like you guys have been. So there is a defined pathway. You cut it at the place where you introduce a new component and the rest of the pathway is being taken care of of the new, by the new component.
Is that correct?
Amanda: I wouldn’t even call it a new component. I would have a replacement of a component. [00:13:00] So the viewer as a component, our viewer replaced Hamamatsu’s viewer. And so we had their SDK. Um, to help us with ingestion of those images. And then of course, our software needed to make sure that we’ve returned the image at the same fidelity, at, you know, a good fidelity.
So that basically if I’m looking at the image in the Hamamatsu viewer, and I’m looking at the image in the Concentric APDX viewer, I can’t tell the difference. The image quality is the same.
Aleks: So I want to emphasize something because there are more cleared systems at the moment. But you guys were one of the first partnerships. Where you had cleared scanner with everything cleared, uh, and then a different component that partnered with this scanner that kind of opened this way of working, uh, for the other partners. I see it as the COVID heritage that we had in the digital pathology space where, uh, we got this, um, provisionary this emergency allowance to actually use whatever, whatever is working.
If you as a [00:14:00] pathologist decide it is diagnostic quality, feel free to help patients. We are in an emergency state and I was kind of like hoping that they’re never going to take it back. But, uh, this is just my wishful thinking of a digital pathology enthusiast, um, at some point they did, they gave a cutoff and those who are serious about it and who kind of already knew, okay, we want to do it in a compliant way, regardless how long this emergency lasts, were those who, who got their systems cleared the fastest.
Scott: Yes. So the, if you workflow, as we just stated, we’re going to start with the scanner. So the scanner is doing its thing. As we move forward in the workflow, it’s the viewing is the next step and working with Proscia. Proscia offers the IMS, the next step, which includes the viewer. It makes perfect sense to us where Proscia would come to us and said, we want to work with you so that we can submit our software to continue down that digital pathway workflow. The nice thing from our standpoint is that we’re working with Proscia and we’re not actually offering an IMS ourselves because we’re not IMS. We’re not an IMS company. We build scanners very well, so we know we knew Proscia had a fantastic interface in an IMS, so let them do what they need to do.
To, again, go down the digital path pathology workflow. We go to Proscia. Proscia actually develop their viewer integrated into their IMS. Then the next step would actually be from there integrating into an LIS. That’s the workflow. And we are open. We’re open for other vendors to come to us and say, Hey, we’ll, we’ll willing to work with you.
Would you help us? And sure, because that now offers different choices to the marketplace for digital pathology solutions and for software packages to work best. For individual customers needs, which is wonderful.
Amanda: And I think the point you made earlier around [00:16:00] COVID and having that enforcement discretion, hell, a lot of laboratories using equipment, hardware, software, that was not cleared.
And without any issue. I mean, I know we were, I was a part of a meeting with the FDA shortly after, right, as the enforcement discretion was ending to talk about the data that was collected. Um, as part of the Digital Pathology Association DPA and, you know, collected a lot of data to show that we didn’t see adverse events, uh, events or anything like
Aleks: That was, like, the basis why I was so hopeful.
Maybe we’re gonna have enough data that they would just say, it’s all good.
Scott: Yes.
Amanda: I mean, to some extent, customers were using what they already had. So they might already had scanners that were, um, not clear, but they were using them for research purposes or what have, or at the time, still a lot of folks were focused on a laboratory developed test route, um, laboratories, the LBT route.
I think that sort of opened up the door and people were like, we want, we, We’ve already kind of done this in practice and you can see that there are different components that can work together very successfully. And [00:17:00] so I knew from Proscia’s to seeing a point there, so our Hamamatsu, we interact with lots of different scanners and we have IVR clearance with Hamamatsu, including other, um, vendors as well.
And we do see it as like, you know, laboratories may want to have different scanners from different vendors and just a central interface, you know, a viewer interface and IMS, you don’t want to have multiple. Those, you want a central software and we want to be able to provide that, you know, open platform for them.
So I think the COVID thing sort of got, maybe pushed that a little bit more because people are already like we’ve been using lots of things that aren’t a part of like a, a connected system, but there are advantages to the type of clearance of a clerk system and the type of the partnership we have with Hamamatsu.
There are advantages that well than doing the sort of piecemeal component, you know, pick your own version of things.
Aleks: Yes, let’s talk about those advantages. Before we talk about those advantages, I want to ask a logistic question. Hamamatsu, you guys had to put the scanners in three different locations. Um, the Proscia team, did you guys just need to [00:18:00] get a certain amount of digital slides because you were comparing digital to digital, whereas Hamamatsu was doing glass to digital?
At all? What were, what were the different logistics between your, um, study designs for the submission?
Amanda: They’re slightly different. Um, you know, for our standpoint, we did a technical performance assessment testing. We also hear people refer to it as TPA for short. And that’s basically, um, their guidelines that the FDA have put out on performing that kind of, um, testing for digital pathology that are fairly prescriptive.
And that’s doing a delta E comparison between images. So what’s the difference in image fidelity between the two? So we did that kind of testing. And then we also did similar to, um, what Hamamatsu did is a clinical study…
Aleks: The diagnostics…
Amanda: …but we had three different sites, but for us. We didn’t have to have the scanner at each one of the sites.
We just needed to have all of our slides that were using the clinical study scanned, and we can do that in a central location. But it’s interesting enough that during the study,[00:19:00] we actually felt some of the same logistics challenges that our labs face today with trying to transfer and do logistics of physical slides across these different.
Uh, participate in clinical sites, you know, once we got to the digital arm, you know, digital arm of the study, it was so much easier because, you know, you could basically just quickly distribute, you know, digital images for review. And, you know, it was instantaneous, you know, it’s paralyzed things a bit. So that was, I thought that was a nice take home where like, we basically proved our own value internally in our own practices with the, with the clinical study.
Aleks: Yeah, that’s what the, one of the advantages of digital pathology, like the most obvious one.
