I can just share my slides, give me a minute. So, just to give an overview, today we'll just talk about hereditary cancer syndromes as well as predisposition genes, how we offer germline genetic testing, how we interpret a test result, as well as maybe run through some case studies. And I know that I'm surrounded by experts here as well as very learned audience, and predominantly everybody does gynecology oncology here. So, we really just talk about ovarian and endometrial cancer in the short time span that we have. So, once again, only a minority of cancers are actually due to hereditary cancer syndromes, and this applies to most solid tumor types, including gynecological cancers. And by hereditary cancer syndromes, we really talk about that five to 10% who have pathogenic germline genetic variants. But majority of patients actually will have sporadic cancer and not familiar hereditary cancer. But just to jog the memory of those of us who may not remember, sporadic cancers actually do not have anything abnormal at germline, and actually the mutations that they have are in the affected tissue, such as the ovarian cancer cell, and cannot be passed down from parent to child. But for hereditary cancers, you would know that the germline mutations actually occur in primordial tissue, like sperm or eggs, and can be passed down from generation to generation. I think this is really key to understanding mutation testing. So how about ovarian cancer? Cancer predisposition genes common. We know that based on large meta-analysis, more than 10%, depending on which paper you look at, will have a pathogenic variant, with BRCA1 being most predominant, followed by BRCA2. But there's a smattering of other cancer genes, particularly in the homologous recombination pathway that will be implicated. And I show you here a list of candidate genes, particularly for hereditary breast ovarian cancers, generally not prevalent in the general population, but however, are significant because they increase the risk of ovarian cancer from 1.1% to as high as 58% for BRCA1. You can see that for BRCA1, BRCA2, PELV2, and ATM, there are associated cancers like pancreas, breast, as well as other tumors that we will need to think about. But however, we also do have some ovarian cancer-specific genes, like BRCA1, RAP51C, and D, that are also quite important because they increase the lifetime risk to about 15% to 20%. But Lynch syndrome also has a role to play in ovarian cancer, and you can see that there are different risks accorded, depending on the different types of mismatched gene deficiency that one would have. But I would like to caution that germline genetic mutations can be detected without any family history, and therefore, universal testing of ovarian cancer is actually advised as per guidelines. And here you can see the latest NCCN genetic testing guidelines, you know, in March 2024, which actually says that actually for ovarian cancers, we recommend testing in ATM, BRCA1, 2, BRCA1, Lynch syndrome, RAP51C, and D. And in addition to informing on cancer risk and prevention, we now can think about PARP inhibitor therapy that you all would probably be very familiar with, as well as checkpoint inhibitor treatment. But epithelial ovarian cancers include fallopian tube and primary peritoneal cancers, both serious, non-serious, and carcinosarcomas, except does not include borderline tumors. And you can see from the Seattle group, actually that, you know, serious cancers predominate, but you can also see that in carcinosarcomas there will be a proportion with BRCA1, 2 mutation. Now, very broadly, and depending on which papers you read, the spectrum of risks for different cancers will be very different according to BRCA1 and 2. But suffice to say, BRCA1 is generally younger, BRCA2 is generally later, BRCA1 has a higher cancer risk, BRCA2 will have a higher, lower cancer risk, but actually predominate for male cancer. So you can see prostate cancer risk can be very high. Similarly for male breast cancer. So when somebody is born with a BRCA1, 2 mutation, we'll offer them enhanced cancer surveillance, and this will include MRI breasts from 25 for women. And of course, talking about ovarian cancer screening, which has not been shown to affect cancer mortality, but risk-reducing surgery for the breasts as well as ovaries can be beneficial in reducing cancer incidents. For males, we actually start chest wall examination at 35, and we do prostate cancer screening as well. But we want you to know that BRCA1, 2 was a model gene that we based all our cancer prevention techniques on, but there are also gene-specific management guidelines for non-BRCA1, 2 cancer prevention genes. I think you should know if you are gonna offer testing. So how about endometrial cancer? Again, endometrial cancer is also a cancer that can be associated with Lynch syndrome, and it's one of the more prominent cancers. You can see that only about 3% of endometrial cancers will ultimately be diagnosed with Lynch syndrome, but it is significant because it brings forward the average age of diagnosis to the 40s and 50s and increases the risk from 3.1% to more than 50%, depending on the gene. And it's interesting to know that there's actually a genotype-phenotype correlation with MLH1 and MSH2 having the highest risk, followed by MSH6 and PMS2, which has a much lower risk. So that, again, in accordance with family history, can allow you to appreciate the risk that a patient will have, as well as the other associated cancers in the gut, as well as in the genitourinary syndrome. But we also know that P10-associated Cowden syndrome is a very rare syndrome, but can be significant in increasing one's risk of endometrial cancer, along with breast, colorectal, thyroid, renal, and melanoma, but it's exceedingly rare. So universal tumor testing has actually been found to be cost-effective for endometrial cancer, just like colorectal cancer, with MMR-IHC and microsatellite instability testing as the first cut. For those who may not be familiar, almost a third will ultimately be abnormal. And then out of the abnormal ones, about 29%, based on a meta-analysis, will have Lynch syndrome on germline testing. So it makes a very nice reflex kind of pathway to do testing for those who may not be familiar. Now Lynch syndrome, again, has many cancers. Previously, we used to call it HMPCC, but we know that colon cancer, while highest risk, is no longer the only cancer associated. There will be stomach, urinary tract, right, endometrial, prostate, as well as ovarian cancers, and the risk can be high, you know, which is why it's really important. And for those who are found to have Lynch syndrome, we start screening colonoscopies instead of age 50, we start from age 25, instead of every five to 10 years, we start, we do it every one to two years. We know that endometrial blind biopsies can be offered. It doesn't change survival, but however, you know, it does help to pick up endometrial cancer, hopefully at an earlier stage in a higher risk population. Now there are clinical criteria to guide suspicion of Lynch syndrome. We talked about MSTEM 1, 2 criteria, the 3, 2, 1 rule that I talked about, but as you can imagine, the sensitivity is really low. The specificity, however, can be very high for this rare occurrence, and people who fulfill the MSTEM 1 and 2 criteria will be managed as Lynch syndrome carriers, regardless of genetic test results. So even if you test negative, we will manage you as if you have Lynch syndrome. However, we have more relaxed criteria like the revised Bodesta guidelines, largely based on colorectal cancers, but endometrial cancer can be used to, you know, substitute colorectal cancers in this criteria, which broadens sensitivity, you know, and has quite good specificity. But having said that, we know that clinical criteria for Lynch syndrome will miss up to 50%, and with the advent of genetic testing technology, we now combine both. So the NCCN guidelines, as you can see, has a large chunk of words that really talk about using clinical criteria that needs family history. But however, with universal testing, we now know that anyone with MMR deficiency or MSII high status can then be eligible for genetic testing. But however, as somebody who runs the cancer genetic clinic, I cannot stress more than enough that family history can really guide us as well. There's still a role, right? Because we know there's overlap between multiple cancers and causative genes, so it would be important. Guidelines for genetic testing have become lax because now, instead of having a, needing to spend several thousand dollars, perhaps 700 may be enough to help test a patient. And also, we know that genetic testing is important because we have therapeutic indications, be it the usage of PARP inhibitors or checkpoint inhibitor therapy that's now tumor agnostic, as well as the usage of risk-reducing surgery. But family history helps us make sense of the genetic test indication, interpret tests in the correct context, and also helps us counsel patients better, especially if they're confused. So how do we do a family history? I mean, this is Genetics 101. It's really not about just yourself. It's about your children. It's about your siblings. It's about your parents and your parents' siblings, as well as nieces and nephews, you know, as well as your grandparents. And remind yourself that, you know, you need both maternal and paternal information. You need to know whether there is genetic contribution. But a lot of times, in the limited time we have for clinic, you know, we personally ask them whether they have cancer, and they will say the family has no diseases. So what sort of use does this family history tell us? Now, what would be a good family history? Again, it's really about having good amount of data, tree generations, the patient's name and date, as well as identity, the age of cancer diagnosis, the type and histology. Now, why? Because we know certain tumor subtypes can be associated. For example, Lynch syndrome may be lower segment, may be high grade, may be endometrioid, may be LVSI positive. We want to know the current age and disease age of relatives, because if you have less living relatives, particularly female beyond the age of 50, we know that the likelihood of observing gyne cancers in the family can be very low. We also want to know relevant medical and surgical history, because if there is a lot of THBSOs being done for benign conditions like fibroids, you can reduce the observation of gynecological cancers within the family. We also want to know whether there's identical twins, because they can have near identical genetic risk. Also adoption status, because they don't add genetic information, and race and ethnicity, because we know certain ancestries like the Ashkenazi Druze have a much higher BRCA1-2 carrier rates, right? But when we draw a family tree, are we even certain the information is correct? So adjudication is important. Now I tell you the same patient now, but now I've keyed in all the information I've gotten from taking the family history. So this is Anna, who's cancer-free, she's Chinese, and can see there's a family history of endometrial lung cancer, colon cancer, and ovarian cancer across both sides of the family. Does she have a strong family history? Let's think about it, right? But how do we decide whether it's hereditary? Well, we need to first ask ourselves, is the personal and family history suspicious for hereditary syndrome? Will we advise genetic testing? What are the candidate genes considered in testing? And who would be the ideal index candidate to start testing? So in Anna's family history, you can see it's suspicious for Lynch syndrome, because there's a family history in the paternal side for young-onset endometrial cancer and prostate cancer. And who would I test? Would I test Anna? No, I would recommend Anna's sister with young-onset endometrial cancer to undergo genetic counseling and consider genetic testing. So which genes will we test for? It can be very straightforward by testing hereditary breast ovarian cancer genes and Lynch syndrome genes, but depending on your clinical suspicion, you might need to add P10 for CALDEN, TP53 for Li-Farmini, SDK11 for Pute-Jaegers, POTE for Proofreading Syndrome. So it really depends on familial risk assessment. There are many providers for germline testing. We could use Myriad. I'm sure many of you are familiar with the HRD testing, but they have a Myris hereditary profile. We can just test for Lynch syndrome alone. We can test for breast and Lynch, but we can also extend to increase ovarian cancer genes. We can do 70 over genes to look for more broader genetic implications, or we can go beyond 100. But for those who may not be familiar, do you even know what the genes mean, right? So before we actually do gene testing, we need to do counseling, and counseling really is to assess understanding, to let us know what the motivation for testing is, to educate patients on hereditary cancers, as well as what we can do to help them, as well as the risks, benefits, and implications of testing, as well as limitations, you know, and how this would impact on the family. So the genetic truth can be distressing, because if you get a positive result, it means that you have a pathogenic mutation that increases the risk of cancer. One can get depression, guilt, negative feelings. If you have a negative result with no variancy, we want to make sure there's no false sense of reassurance given. And in some people with variants of uncertain significance, it can be confusing. And we mitigate these issues with pre and post-test counseling. So variant classification by the American College of Medical Genetics is in five classes, can see that pathogenic and likely pathogenic will be positive, benign and likely benign will be negative. But really, there's this whole chunk called variant of uncertain significance. We know that this variant of uncertain significance means that the chance of it being potentially pathogenic can be as low as 5% to 95%. But generally speaking, it's not actionable. So you can see that in this study we've done in breast cancer patients, we can, you know, get a diagnostic yield that's pretty good with BRCA1,2 testing, but we can increase that diagnostic yield at the expense of getting more VUS if we increase to 15 genes, and increasing to 49 genes increases it up to 35%, which is really acceptable, as long as we know what to tell the patient. But think about those who go on to more than a hundred genes, you potentially will get more and more VUSs. But what do we look at in a genetic germline test report, which you may see in your clinic? One thing is don't jump to the result, make sure that it's the correct patient, make sure that the lab is correct, the methodology is actually sound. What is the gene panel and how updated is this test? Remind ourselves what the indication for testing is, because a lot of times we may not even know why the patient went for testing. Look at the variant, is there functional impact? Does it correlate with the patient's diagnosis? So again, when we see a report, we look at the lab, we look at the patient details, look at a sample, we look at a report date. But more importantly, in the methods, we need to see, in this case, it's next generation sequencing, it looks at targeted sequence and a flanking introns. We saw that large deletions and duplications are accounted for, and that if they see pathogenic variants, they subject them to confirmatory testing. And then we look at the gene list, what did we test for? In this gene panel, we saw that we had hereditary colon cancer genes, we had hereditary breast or brain cancer genes, and we also test for less common hereditary cancer syndromes. Then we go to the meat of it. So I thought the best way is to really tell you some case examples. This is Madam A, I've made it easy for you. Madam A has got FIGO2 endometrial cancer with tumor loss of PMS2 and MLH1. Probably not uncommon in your clinical practice, but essentially, MMR deficient endometrial cancer. Even in the absence of family history, we know that there's a slightly under 30% chance of getting a germline genetic variant, so we should offer germline testing as a reflex. In her case, it tested negative. Are we surprised? I wouldn't say we're surprised, because the majority will be sporadic. But what do we tell the patient? We tell the patient that there were no germline genetic tests seen in the candidate genes tested. It does not exclude hereditary cancer from untested or unknown genes if you suspect so. But majority of endometrial cancers we know are probably not hereditary due to strong genetic factors. But we also do know from epidemiological data, the relative risk is about 1.8 to 2 for patients, for relatives with endometrial cancer. Now, predictive testing will not be offered and cannot be offered, and cancer screening, however, has to be tailored. While there's no good population screening for endometrial cancer, we do know that we can control cardiovascular risk factors, and this is a learning moment to really do population cancer screening for cancers that can be screened. How about Madam B? Madam B has ovarian cancer. This is FIGO3 hybrid, but she got it at a really elderly age at 71. She's now 73 and cancer free. Should we offer testing? In this case, there's hybrid serous cancer and a maternal family history of breast cancer. We should exclude hereditary breast ovarian cancer syndrome. What does she test for? She had variant testing and was found to have a variant of uncertain significance in BRCA1. Then we look at the variant. Why is it a variant of uncertain significance? Well, it's a duplication mutation that results in frameshift. That's suspicious, but however, you know, the frequency data is actually not high. And also, while we have seen patients with colorectal cancer, ovarian cancer, you know, the public databases do not have sufficient data to support that this is pathogenic. Neither do we have functional data. So what do we tell Madam B's family? We say that we didn't find any pathogenic germline mutations. The variant is not actionable, but follow-up is required. We know that about 30% of BOSs will be reclassified within two to three years, and out of this 30%, about 90% will be downgraded to benign. So therefore, it is important that we give these patients this reassurance, but it does not exclude hereditary cancer from untested or unknown genes. We know that the kindred will be at high risk of cancer, but we will not offer predictive testing using this variant. And cancer screening can be tailored, although perhaps the data for ovarian cancer screening is still controversial. Let's not forget population guidelines for cancer screening as well, particularly for breast cancer in this family. Madam C, again, has got high-grade serious ovarian cancer, and her sister has got ovarian cancer. In this case, it's an ovarian cancer family. Will we offer genetic testing? Yes, we will. So in her case, we did germline testing. We saw a pathogenic variant in BRCA2. Well, this germline deletion in BRCA2 resulted in a truncating mutation and can explain the personal history and family history of ovarian cancer. So all kindred, in the absence of genetic testing, must be managed as mutation carriers. If you do predictive testing for them, if they're found to have the mutation, then they're at high risk of breast and ovarian cancers that can be offered enhanced surveillance and risk-reducing surgery. But if you test the children and they don't have the mutation, then we call them true negative. Their risk of cancer will be average. They do not need enhanced surveillance or risk-reducing surgery, and population screening guidelines can prevail. Now, this is Madam D, which is not unsimilar to the patient presented at the case report. You can see that it's a busy family tree, but if I summarize for you, she has got young onset endometrial cancer at 37. She has three generations of colon, endometrial, as well as breast cancer. This is Amsterdam 2 criteria. So therefore, as mentioned, all her kindred should be counted as Lynch syndrome carriers, but will we offer testing? We will. Just like your patient today, once we find a pathogenic mutation that is causative, we can then offer predictive testing to the family members in the maternal tree. And in so doing, we hope to relieve 50% of them from unnecessary screening. Last but not least, I have Madam E, who has got endometrioid ovarian cancer, age 45, as well as a grandmother with endometrial cancer. In her case, we would want to rule out both breast ovarian cancer syndrome and Lynch syndrome. So you sort of see this common theme that's going through all the cases I'm trying to present, but she comes to your clinic with her HRD test instead. So this is Myriad MyChoice, which many of you may have already used. She has got Myriad HRD positive score, but however, the tumor mutational BRCA12 status is negative. Does this mean we don't have to test her anymore, since she already has some form of genetic testing? Well, we went on to do germline testing. She had a pathogenic mutation in PelB2, which we're not entirely surprised by. What this means is that this variant can explain some of the history in the family. And in the absence of genetic information, we should all manage the family members as PelB2 mutation carriers, but we can offer them predictive testing. Now, why am I bringing up this case? I'm bringing up this case because I just want to let you remind everybody that somatic tumor testing is not equivalent to germline testing, because most tumor tests do not state whether the identified mutation is somatic or germline in origin, and most of the time there's no paired testing. The likelihood of mutations detected in tumor tissue being germline origin really differ by different genes. Looking at this Memorial Sloan-Kettering study, you can see that across different genes, if it's BRCA1 mutated in the tissue, then the likelihood of it being germline, relatively high. If it's Lynch syndrome, moderately high. If it's TP53, extremely low. But of course, if you have a family history that's suspicious, it can sort of help you tweak your understanding about whether this could be germline or not. So mainstreaming genetic testing for gynecology patients at our center has now been done for epithelial ovarian cancer and MMR deficient endometrial cancer. The main question is really to ask yourself, what's the intent? Is there an urgency? Is the family history suspicious? So our gynecologists do brief testing for patients, and if negative, they counsel them, and if pathogenic, they refer them to our clinic. If varying of uncertain significance, they also, you know, can send them to our clinic through a fast-track referral process. And this is made possible mainly through the use of clear guidelines, easy communication, as well as smaller panels to reduce the VUS rates. So how did we do this? We have the NCIS Cancer Genetics Workshop, where I had the privilege of meeting Dr. Zi, and we have done this since 2018, right, and we empower non-genetics healthcare professionals to provide genetic counseling. It was mainly in-person, but we pivoted to virtual platforms during COVID. We have didactic lessons, as well as hands-on interactions. Many people were put in the hot seat. We had a Certificate of Competency issued after some assessment, and we tailored it to different groups, and we've added on a somatic component in ovarian, as well as prostate cancers in 2021, and we've been doing this workshop about one to two times a year. So to date, we've actually trained more than 100 healthcare professionals in the region, and really is to create a community of practice, and, you know, in our next sessions in July 2024, and I believe there are quite a few from Hong Kong who will be joining us in Singapore this year. So in summary, I hope I've convinced you that genetic testing is recommended for all epithelial ovarian cancers, and a significant proportion of endometrial cancer, for both familial and therapeutic reasons. The demand for genetic testing will only increase, so let's not believe that it will go away. It's really important for gynae oncologists to have an understanding of principles of familial risk assessment, as well as genetic testing, but again, not to sound like a nag, before testing, counsel, after testing, counsel, and integrating cancer genetics into mainstream pathways can only be useful and an asset, particularly in challenging situations. So again, thank you for your time. This is my email. Happy to collaborate. Thank you.

hereditary cancer syndromes

gynecological cancers

BRCA1 BRCA2

Lynch syndrome

genetic testing

genetic counseling