Um, so I thought we'll talk today about HER2 assessment and really we'll talk today from the pathologist's perspective so you will understand the complexity of the testing and how really to read the results. When you look in the, I'll just minimize this, okay, so the history of HER2 starts in 1998 when trastuzumab was first time approved for the breast cancer and now it's a standard of care that HER2 assessment is done for every single breast cancer. And it's up until 2018 and you can see significant gap when trastuzumab was introduced into gynecological and GI malignancy and you can see here it's gastric and endometrial carcinoma. And in 2019 NCCN endorses trastuzumab for endometrial serous carcinoma. Obviously now the story continues with the conjugate but this is a little bit different and we'll talk about that in just a moment. So in the landmark randomized phase two clinical trial in 2018 it was demonstrated that trastuzumab in combination with carboplatin and paclitaxel significantly improves progression-free and overall survival for HER2 positive advanced stage and recurrent endometrial serous carcinoma compared to just chemotherapy alone. And that's why NCCN guidelines recommend that addition of trastuzumab to standard chemotherapy. And that's basically the reason why we're talking about this right now. And then earlier this year, as you know, FDA grants accelerated approval for Druxacan for unresectable metastatic HER2 positive solid tumors and endometrial tumors are included as well as other gynecological sites, ovary and cervix. So HER2 protein is encoded on chromosome 17. It's a therapeutic target. There are site-specific protocols for HER2 assessment with criteria for IHC and in-situ hybridization with significant variation in interpretation. So that you as a clinician need to keep in mind and ask your pathologist, what protocol are they using? Because there is a lot of mixed up promising data in some sites, but no universal testing in all gynecological malignancies. And really addressing heterogeneity of staining both immunohistochemistry and ISH and details of testing are still in progress. Just to remind everybody, HER2 is one of the proteins of 4-protein family and it has an extracellular domain and the work of EGFR and HER2 actually influences proliferation and differentiation and obviously goes into survival and invasion. Sites for HER2 assessment, I mentioned endometrium and we assess it, as most of you know, endometrial serous carcinoma, but also we do assessment on all p53 aberrant endometrial carcinomas. Up to date, there were no HER2 positive cases in p53, it's in p53 normal tumors. And we assess mucinous carcinomas of the ovary, cervical adenocarcinomas, and we do assessment of budget disease of the vulva, but it's more really for diagnosis rather than prognostically. There are a lot of questions that one can consider, it's when to test, what tissue to test, when to do retesting, and many of us do upfront testing and others only do at the request of the clinician at the time of the recurrence. There are a number of HER2 guidelines and they vary significantly, the main one is for breast, another one is for gastric and endometrium, and again, it's not specified what tissue to test. Just to let everybody know, here on the slide you see back-to-back all different sites for HER2 testing. So for breast, the cutoff is 10% for positive HER2, for gastric 10%, and for endometrial serous carcinoma is 30%. For fish, there are a little bit slightly different guidelines for breast, they're more extended, although the HER2-17 ratio should be over 2 regardless of the site. Then there is a significant discrepancy between number of cells required for amplification in gastric and endometrial carcinoma. In gastric, 5 cells would be enough, where in endometrium we're talking about 20 cells, and that's overall testing criteria for HER2 endometrial carcinoma. Recently this year, one of the papers published from our institution showed significant heterogeneity in up to 40% of endometrial carcinomas, and you can see here, so this on the first slide, this is endometrial serous carcinoma, and only a fraction of this tumor exhibits HER2 positivity, and it's on the pathologist to decide whether it's 30% or less than 30%, and if it's less than 30%, the tissue will go into in-situ hybridization. And now keep in mind, if you will do in-situ hybridization just in this area, it most likely will be positive because all you need is 20 cells. And here are other tissue fragments for you to demonstrate that heterogeneity. You see significant portion of tumor. This is a tumor on the left-hand side, and this is immunohistochemistry, and only about, I would say on this slide, less than 10% of the tumor is positive. And look at this complex heterogeneity of staining on that tumor. It shows different patterns from totally negative to plus one, plus two, and then a positive one. Another picture to demonstrate similar point. So here you see, basically when we call something IHC negative, it has to be completely negative according to the current guidelines. You can have a very, very faint staining, but because of the new drug, most pathologists will call IHC score zero when they see almost no staining. And then IHC score one plus, when you see that faint staining, and it's quite different from breast criteria, you can see basolateral, not necessarily full membrane staining. And then going to plus two, and those cases will go to FISH, two plus remember can be either amplified or not, so it can be overexpressed or not. And three plus, and you have that very diffused positive cells. So there is a variability of immunohistochemistry, HER2 assessment. It's a standard in many academic centers across North America, but we don't know the global extent of the testing. Unfortunately, previously published papers show a variability of overexpression for four to 69%, which is totally unacceptable, as you can imagine. And again, in our institution, we did a focus study of 180 patients, and we found 28% of tumors were overexpressing HER2. And another point to keep in mind from Natalia Buza, who was on all the trials with Dr. Feder, the overall agreement, the overall pathologist agreement is about 72.3%. It's in our assessment of HER2. In general, I just want to give a little bit of background to immunohistochemistry. What is the problem with biomarker assessment? There is really no pre-analytical standards, and pre-analytical is everything that happens before the tissue is cut. So we're talking about tissue handling, microtomy, quality assurance checks, preparation for molecular testing. There are also no post-analytical standards, and that's the biggest issue, because there is really no standard reporting. So you will get one report from me, another language of reporting from another pathologist, and so on and so forth. And then final summary and checks with all the studies are not necessarily included. There are multiple professionals involved. There are surgical pathologists, molecular pathologists, scientists, and sometimes those reports really not talking to each other. We don't have a clear guidance on turnaround time for biomarkers at that present time. Just to say very quickly, the problem with pre-analytical, you can think, well, why do I need to know that? Because about 80% of diagnostic errors are related to pre-analytics. So if you're not doing a good job there, we will not get a meaningful result. Very important tissue sectioning, and there is, again, no guidelines on the thickness of section. And we understand the quality of section, but again, there is no clear guidelines. And remember, the tissue needs to be properly fixed in the formalin. When we talk about FISH and in situ hybridization, that's where the expertise of the pathologist comes into significant play. And that's what I want really to emphasize for this session. A lot of the times, FISH is done either by molecular pathologists who do not see H&E slide and don't see immunochemistry and surgical pathology, just make a little circle on the slide for them to select the area. And then very rarely, the in situ hybridization is done by the same surgical pathologist. So they will do a whole section assessment, and they will check entire slides who might get a different result. And that was actually shown in one of the studies that I quoted before. You really need to have an experienced readers dedicated to that site with the understanding that there is significant heterogeneity. And FISH criteria, I will skip. We just talked about that. So this is a picture of FISH. And the reason that I'm showing it to you, that means the pathologist is sitting in a dark room looking at those slides. So they don't know where they are. So if they didn't see the H&E slide, they didn't see the immunoslide, very difficult to know what exactly you're looking at when you are sitting in that dark field. So you see cells, they are amplified because all these red signals, it's HER2. But if you don't have a context, very difficult to be certain about tumor versus not tumor. So as I mentioned to you, there is significant heterogeneity of HER2 protein in endometrial carcinoma, and up to 40% of samples show heterogeneity, and sometimes it's a complex heterogeneity. So if we have a 20 cells or so of heterogeneity at site, we'll probably okay to decide whether it's positive or negative. But we don't know what to do with the cases when it's a scattered cells showing increased signal. So this is not addressed in current guidelines. A lot of us will mention that in the report, but again, what to do with this patient is not clear clinically. Another modality for you to consider is NGS, and it's getting very popular because there are a number of machines now that are affordable. So it's becoming very easy and not labor intensive in the labs. So labs are buying little machines and running NGS, but unfortunately, there is significant discordance between IHC fish and NGS for HER2 in up to 24% of the cases, and it can be related to quality of the tissue, low tumor content, small size of the region that is amplified. Again, because the tissue is heterogeneous, so you have to land on the region that is a region of interest, right? And low level amplification. That's what I wanted to tell you. And it's just maybe to talk to your local pathologist to explore what they do, what methodology are they using, and how can you really have a comprehensive report at the end of all the methodologies that you're getting, how can you have one single report that explains to you all the biomarkers on one page.

HER2 assessment

cancer treatment

pathologist expertise

biomarker data

treatment outcomes