At the National Society of Genetic Counselors (NSGC) 2020 Virtual Annual Conference, Ambry discussed +RNAinsight® and the CARE Program™. The following addresses some of the questions we received.
In case you missed this webinar, you can view a recording below.
+RNAinsight®: Paired RNA and DNA Genetic Testing for Hereditary Cancer Risk
RNA-Related Questions
| Questions | Answers |
|---|---|
| Is there any progress toward RNA analysis from saliva instead of blood? Will it ever even be possible? | Isolating RNA from saliva is challenging because over 80% of salivary RNA is bacterial. To determine if RNA data derived from saliva is as reliable, conducting further investigations and validation studies are required. |
| Is there any insight into when we should recommend updated testing with RNA analysis for patients whose past DNA-only test results were negative despite having a family/clinical history of a hereditary cancer syndrome? | As you mentioned, it may be important to consider updated testing with RNA for patients with a family/clinical history strongly suggestive of hereditary cancer. However, we have observed clinically relevant RNA findings even in the absence of a strong family history and in genes that have moderate penetrance or common phenotypes, such as breast cancer. |
| How does identification of an intronic variant impact TAT? | If an abnormal transcript is identified, we sequence the intronic DNA coordinate using Sanger sequencing to identify a causal variant. Usually this process takes no more than a few days. The vast majority of cases are still reported within our standard TAT of 14-21 days. |
| Is there any data on the percentage of cases where RNA was able to find a P/LP variant when DNA was negative? | Our preliminary data shows that 1 in 43 positive cases were impacted by +RNAinsight. Of those impacted, 30% would have received a negative result from DNA-only testing, and 70% would have received an inconclusive result. |
| Why do DNA probes not sequence introns as well as exons? | This is primarily due to the size of introns and lack of conservation. Overall, many probes would be needed to sequence these regions of the genome. More importantly, without functional data, variants identified in the introns would most likely be classified as VUS. |
| Would offering RNAseq on a reflex (clinically actionable) basis be more appropriate from a utilization management perspective instead of concurrent testing? | We previously utilized retrospective RNA genetic testing for specific variants in Ambry's Translational Genomics (ATG) lab. With this approach, we observed two critical limitations. First, deep intronic pathogenic/likely pathogenic variants that were clinically actionable still remained undetected. Second, there was a 90% lost to follow-up rate for patients who were offered additional RNA studies due to the need to send in an additional sample. Therefore, it is more advantageous to offer RNA testing concurrently with DNA testing. |
| Do you include information from RNA-seq findings on ClinVar descriptions of classification? | Yes, in alignment with our long history of data sharing, variants impacted by RNA genetic testing are submitted to ClinVar, and we include a brief summary of relevant data. |
| These examples presented are in the APC gene. Are there certain genes that are more likely to have these deep intronic mutations? | The data is still evolving. Thus far, we have observed deep intronic variants in several tumor suppressor genes beyond APC, including ATM, NF1, BRCA1/2, and in the MMR genes. |
CARE Program Questions
| Questions | Answers |
|---|---|
| In considering these new service delivery models, what are your thoughts about how they may impact access to services? What about the widening gap between those who can use and/or have access to these technologies/devices and those who cannot/do not? | The goal of these types of programs is to increase access and decrease health disparities. We work to remove many common barriers patients have with accessing genetic testing by implementing the following:
|
| How do you know if a patient read the consent or pre-test education material? | It's documented in our back end system. However, it's important to point out that consent for genetic testing happens with the medical team and not through the virtual assistant. |
| Is there a way to assess what the patient understood about the pretest process, like a knowledge check, for example? I could see patients skipping through the prompts and then not understanding what testing would mean. | Obtaining informed consent for genetic testing happens with the patient's medical team and not through the virtual assistant. A knowledge check has not been implemented within the chatbot at this time. Medical staff answer questions and/or address concerns in-person prior to testing. |
| If patients have questions that can't be answered by a chatbot, or if they are unsure about proceeding with genetic testing, how is this handled? | They meet with the medical staff on site to answer questions and/or address concerns. Additionally, optional pre-test counseling is available. |
| Does Ambry conduct the pre-test counseling? Is this only available for certain indications or across multiple oncology specialties? | Pre-test genetic counseling is conducted by a third party GC group. Regarding oncology specialties, the assessment primarily uses NCCN guidelines for HBOC and inherited colorectal cancer syndromes. Though, the chatbot does ask about a range of cancer types. |
| Is a pedigree generated in the pretest session? | This functionality is being added in early 2021. |
| Has there been any research about chatbots and cascade testing, especially in instances of strained family relationships? | Here is one study that was conducted regarding chatbots and cascade testing, although it does not specifically address strained family relationships. Patient assessment of chatbots for the scalable delivery of genetic counseling J Genet Couns. 2019 Dec;28(6):1166-1177. doi: 10.1002/jgc4.1169. Epub 2019 Sep 24. |
| Is there any explanation about different levels of testing? For example,some patients want a large panel while others prefer to limit incidental findings, choosing a more targeted panel. In this setting, how is the type of test to order decided? | The ordering healthcare providers determine which test to order for their patients. CancerNext is the most frequently ordered test, which aligns with overall genetic testing order trends. |
| Do you know why >50% of patients who qualified for testing did not proceed with testing? | We are still collecting and analyzing data related to this important question. Thus far, we have observed that some patients already had genetic testing. Others report disinterest but don't give a specific reason. We plan to look at this more closely to better understand why patients decide not to be tested. |
| Often, family histories are much more complicated than an AI program can assess, and a genetic counselor might find a patient’s history concerning when an AI program does not. How are you educating providers and patients that this technology could miss many patients that need testing? | At this time, the chatbot does not specifically tell patients whether they are flagged as meeting the evaluated criteria. The discussion about meeting criteria occurs with the healthcare providers, who are educated that this assessment may not flag every patient who needs genetic testing, and that further evaluation may be needed. Additionally, the site may choose to refer the patient to a genetic counselor for more complicated histories or scenarios (i.e. patients with a history of heme malignancies or bone marrow transplants). |
| Do you know how this uptake in testing compares to a traditional model? | We are currently exploring research opportunities that will compare two models: referring to a GC vs. the CARE program strategy. |
| Is this virtual assistant available in any other languages? Do you know at what level the literacy questions and education are aimed? | The virtual assistant is currently available in English and Spanish, and we are currently adding languages. Most content is aimed at a 7th to 8th grade reading level, but it varies with detailed explanations of more complex clinical conditions. |
| When was the CARE program first implemented? Where did the 20K patients come from, and how long did it take to accrue that many patients? | The original version of the CARE program was implemented about 3 years ago. However, the data presented from 20k patients was collected over a span of less than six months from multiple clinics. |
| Can you give specific examples of how GCs at institutions should be involved in the implementation of CARE? | GCs can be involved in a variety of ways, from helping set up the entire program to identifying key stakeholders at the institution. The GCs frequently help determine results disclosure preferences, such as whether patients with positive results are seen by a third party GC company or by GCs onsite. They are also involved in discussions about how to manage insurance requirements for pre-test genetic counseling and/or workflows for various clinics at the hospital. |
| Will the CARE program consider family history to help identify other family members at risk? | Every patient who tests positive receives genetic counseling and is informed about the need for cascade testing. |
| If patients with positive results never follow up for genetic counseling, will they eventually receive a paper or electronic copy of their results? | Patients who don't schedule a genetic counseling appointment receive follow up from genetic counseling assistants via phone, email and/or text to make sure they get the care they need. |
| Will the CARE program update with every new iteration of the NCCN guidelines? | Yes, the risk assessment is updated when there are relevant updates to testing guidelines. |
