Developing Story: CureWise Goes Live - The AI Platform Built by Cancer Patients, for Cancer Patients

Last time, we had a front row seat to a vision. This time, that vision has a launch date.

On June 20th, CureWise officially launched, and for every cancer patient who walks into an oncologist's office with ten minutes and no idea what to ask, it couldn't come at a better time.

In this episode of Live From Stage 4, Victoria Goldberg sits down with Steve Brown (founder & CEO of CureWise) and Lisa Booth (metastatic breast cancer advocate and CureWise's patient #2) to talk about what it actually means to have an AI platform that knows your full medical history, speaks your language, and helps you show up to your appointments prepared and empowered.

Also joining us: Beta users Lynda Weatherby (MBC patient and Live From Stage 4  team memberr) and her husband Todd Weatherby (CEO of Siemens Advanta, former VP at AWS), bringing the real-world patient and caregiver perspective to the conversation.

In this episode:

• How CureWise pulls your medical records from multiple institutions automatically
• The Cross-Check feature: asking ChatGPT, Claude, and Gemini the same question simultaneously
• Why 60% of cancer patients have an actionable mutation, but only 15% are being treated for it
• How to prepare for your oncology appointment using AI
• The caregiver experience and what's coming next on the roadmap
• Pricing, the 7-day free trial, and how to get started

Important note: CureWise is not a doctor. It is a powerful tool to help you understand your medical records and have better conversations with your care team. All medical decisions should be made with your physician.

🔗 Sign up at www.curewise.com

Thanks for listening. If you enjoyed the episode, subscribe and leave a review — it really helps. Follow us on social media @livefromstage4 and visit our website at www.livefromstage4.org for show notes and links.

 Your support helps us continue to share important stories and advocate for those living with metastatic breast cancer.

Until next time, take care and keep pushing for progress.

Transcript

SPEAKER_05 0:09

Could your foot cancer be closer than you think? Welcome to Life from Stage 4, where MPC takes center stage as we talk to experts, share inspiring stories, break down signs, and shine the spotlight on what matters most. Because when it comes down to it, the spot, for us and by us, is all about us. Welcome to Life from Stage 4. I'm Victoria Goldburn. Last time we had a front pro seat to a vision. Today, that vision is live. On June 20th, QOWISE officially launched. And for every cancer patient who walks into an oncologist's office with 10 minutes and no idea what to ask, it couldn't come at a better time. Steve Brown and his team built Curewise so you could walk into that room prepared. We are sitting down with a group of people who are truly at the intersection of high-tech innovation and deep personal advocacy. First is Steve Brown, founder and CEO of CureWise. Steve built this AI platform after his own rare cancer was missed by doctors. And now he's making sure other patients have the tools to find the right answers faster. Joining him is Lisa Booth, a powerhouse advocate, an 11-year metastatic breast cancer patient, who was among the very first to use PureWise. In fact, she's patient number two. Lisa is a living example of how having the right data at your fingertips can completely change your outcome. We're also joined by Todd and Linda Witherby. Todd brings a rare combination of tech expertise and personal investment. As a former VP at Amazon Web Services and current CEO of Siemens Advantage, he understands better than most how technology can show up for human health. And Linda is a dear friend and a vital part of our community. She is the one who connected the dots, and her heart for this conversation is really the glue that holds it all together. Welcome back to Developing Story. Curewise is here. But today we're here to talk about the launch and to help us figure out what's going on. My friend Linda Weatherby, and we have a special, special guest, Todd Weatherby, here to tell us what we're doing wrong or what we're doing right. Hi Todd, good to see you. Good to see you. So, see, Lisa, welcome back and congratulations. This is such a big deal. So, how do you guys feel about this?

SPEAKER_03 3:42

We've been using it ourselves for a long time already. So, in some ways, it doesn't feel new for ourselves, but it feels really amazing now that other people can sign up and see what we've been seeing. I'm a year and a half in to cancer, and I still use it all the time. I used it yesterday, the new release version. I used it for about an hour before my doctor's visit yesterday. I've been super lucky and I've had a complete response to the treatment options I figured out on CareWise and got my doctors to do. I still have all these questions like how long do I need to keep doing this therapy? Explain the MRD testing, and did they do it right? And it turns out I learned that my other doctors didn't do it right, and so I have to get it redone. But it's all this stuff because I'm talking to CureWise, learning about it, having a better conversation with my doctor, even a year and a half in, after I feel like, what more could I possibly learn? I'm still learning new things and using it every time I go to my doctor.

SPEAKER_05 4:49

Let's take a moment here. Still just mentioned MRD testing, minimal residual disease testing. It's a highly sensitive blood or bone marrow test that looks for tiny traces of cancer cells remaining after treatment. Think of it as a very powerful microscope for your bloodstream. It can help determine whether your treatment is actually working at a cellular level. And as Steve's story shows, whether it was done correctly matters enormously.

SPEAKER_00 5:20

Lisa, how about you? I am most excited about the new user interface because we've been working with our beta customers and we got feedback and things like that. And so we came up with something that I feel is really intuitive and straightforward. I love that there's another metastatic breast cancer patient in Seattle that's a UX expert, and she helped us design it as we went through it. So it really is working for patients with the patient at front of mind. I love that it shows me its logic as its thinking and it gives me references. And like Steve, I use it all the time as well. I just had a scan where there was an open question of what was going on with my scan, and I had MRD testing, and I was able to analyze those two things together and send an email to my oncologist to say, hey, I think it means this. I'm feeling really comfortable about this, and I think that we should hold on my current chemotherapy for a little bit longer so I can further reduce the side effects I'm having from my chemotherapy. And so I felt I have agency, and I'm so excited that everybody like us can have agency. And I've been using it with a lot of different beta users, and I like that even if you are an early stage person, it helps you understand it. It speaks in simple terms and it describes something. You can actually say to it, I love gardening. Tell me about my cancer as if you were a gardener. And it'll explain your complex cancer or simple cancer in a term that you can understand. And so it's really about creating that very personal experience so that you as a patient understand things and can have a better experience when working with your oncology team.

SPEAKER_05 6:54

You just mentioned it yourself. So you open yourself up for my next question. Let's talk about the beta users. How many beta users have you had?

SPEAKER_00 7:03

We had about 80 beta users.

SPEAKER_05 7:05

Let's talk about their experiences. And Linda, you're one of those people. And it turns out that Todd is also one of those people. So you guys, please chime in and tell us about what your experience is because honestly, I can't wait to get in.

SPEAKER_04 7:20

Yeah. Todd, I'll jump in first. But yeah, I know Lisa and disclaimers. Yeah, no, Lisa's a really prolific MBC advocate. So of course I knew of her, and we're both in Seattle. And so I jumped into the beta testing just a few weeks ago, actually, but it was timely because we're not sure if it's stable disease or not. It has been really helpful in just helping me formulate the right questions going into the most recent appointments. I will say that my husband is very adept in the AR models. He's been using my login to go in and ask his own questions as a caregiver, also as someone with a lot of experience in the tech world. And I think he's got his own story to tell about it. I will share this little nugget. I was on the phone with Lisa last Friday when I read my most recent PET scam, which had confusing stuff in it, really confusing. Let's face it, reading these scam reports is just brutal. And I went right over to the CureWise interpretation. And in addition to having Lisa on the phone to hold my hand as I read the results, what may be very matter-of-fact in a PET scam report, CureWise was able to interpret for me and say, hold on, this requires this and this to confirm. And it kept me from going down a giant rabbit hole on Friday. It was extremely helpful. Todd, do you want to talk about how you've been using it?

SPEAKER_02 8:45

Yes, certainly Linda's journey's been 25 years with cancer, and in the last few years had gotten the roller coaster that all of you have experienced. And so as a caregiver, I'm tracking along and I'm trying to be helpful. I'm trying to understand the science. I'm trying to understand how to help Linda have agency. At least I love that word. You gotta have it together with her and feel like we're somehow in control, or at least we can navigate. And so the idea that CareWise is a navigator, it's a concierge through this crazy experience is really cool. I'll say that for the last six months, I've been using straight GPT myself, trying to emulate what you all have built in a much more robust way. I created my own little council of advisors. I described her situation, I cut and paste her reports from my chart into my GPT chat, and then it tells me I'm running out of time and I'm gonna lose all this history. And I'm thinking, man, if I could only have this longitudinal view of her full history, and I wouldn't have to go fetch all the data myself. If I could only then use my time to have this rich conversation with these tools that give me great insights, wouldn't that be great? And then I turn around and she introduces me to this. I'm a big fan just in a few days of getting familiar with what you've built on the specifics on the tool. I was using Linda's account and tried to invite the caregiver. It generated an invitation from Linda to me, and I couldn't get through it. So my feature feedback is that I had to then set up my own integration with my own healthcare records in order to participate as a caregiver for Linda. And I wasn't there to do that task. I wasn't there to work on my own agency, I was there to participate in hers.

SPEAKER_05 10:30

Let's talk a little bit about that. A caregiver who wants to uh join in, what would be the steps and how do you envision that?

SPEAKER_00 10:39

Let's talk a little bit about that. So we did build the caregiver capability in the sense that you, as a customer, can invite somebody else to have access to your account. They can chat with your medical records and those pieces of the puzzle. Currently, in order to activate that feature, the person has to go through and set up their own account. And our first onboarding flow is really geared towards the customer. And so on our what I call the near next product roadmap, we will be enabling a caregiver onboarding flow so that they don't have to set up your own medical record and those pieces and they're just accessing your medical record, but that will be in a couple of weeks. But it's definitely something that we're actively working on right now because most people have themselves and one to two other people that they partner with. Not a lot of people go to cancer oncology appointments by themselves. And it's as important that the individual and their loved one or their advocate come along with them on that journey. So the intent is to enable that in a frictionless manner. If anybody has any issues until we get this feature launched, they can work with our support team and we will help them move through that process so that they can go ahead and do that while we're waiting for it so that we don't have to be sharing passwords and everybody can have access to both accounts. And Todd, you are going to be our beta tester for this particular function.

SPEAKER_02 11:58

I was interested in the beta part that I saw where you had some notion of personas of clinicians with the faces of fictional people. And like I said, I had created my own little set of advisors and I was intrigued how you did that. It looks like it got parked and isn't uh part of this first release. Maybe you're rethinking that.

SPEAKER_03 12:18

Some things are still in RD.

SPEAKER_02 12:19

Yeah, for sure.

SPEAKER_05 12:20

Or I'm sorry, Steve, I interrupted you because I think what would be helpful for us to actually explain what we're talking about.

SPEAKER_00 12:28

Victoria, why don't we take it from like the start so everybody understands what the process is? Why don't we do that? Good idea. Okay. Yeah. And so for folks that haven't used the CureWise platform, here's how it works. You go ahead and as you onboard, you give CureWise your name so that you have a personal relationship with your agent. And then you're going to say how you want that agent to communicate with you. Do you like things with just the science, or do you need it a little bit softer? Do you like long responses, those types of things? And so we know that people hear information in different manners, and we want the opportunity for the customer to be able to specify how they hear information. The next piece of the process is we connect your medical records. And for some people, that's a scary step. But in order for us to answer your questions in the context of your medical situation, we need to have the medical data to do that. We store your information in a secure manner. We pass it in an anonymized manner to the LLMs. So your data is safe.

SPEAKER_05 13:27

Lisa, explain the LLM models.

SPEAKER_00 13:31

So an LLM is a large language model, and that's Claude, ChatGBT, Gemini, AI, those particular pieces of how things work. But what happens is that for the US customer, because of the CARES Act of 2021, we have the ability to connect directly to your electronic medical record. So my chart, CERN, Flatiron, whatever that system is that your medical provider uses. And so you, through your own two-part authentication, say, I give my permission for my data records to come from my medical facility into a secure spot within CureWise. And then we use that information to create a clinical history for you that you then have the opportunity to review. I liken it to something like if I were going to go get a second opinion and I printed this out, I could take this one piece of paper or two pages in my case to the consulting oncologist, and they could read it and they would know everything about my cancer, the type of cancer when I was diagnosed, any mutations, my biomarkers, the treatments that I've been on, how I've responded, my side effects, any family history, any other complicating factors, et cetera. For me, it would say that I'm HER2, low, ER-positive, metastatic breast cancer, ductal, something or other. I have an ERBB2 activating mutation. I got diagnosed in 2015. I've been on seven lines of treatment, I think. I'm currently on this clinical trial. My circulating tumor cells are zero currently. My last scan said I have two very tiny lesions in my liver and I have some neuropathy. And that's from 1,600 different medical records and documents that I have in my medical journal. It consolidates all of that. So then you can have that intelligent conversation with the agent. So now you land in CureWise. And the first place we land you is a spot that allows you to review that clinical summary. So if there's anything missing from your record or is misunderstood, you have a chance to correct it. It's very much like when you go to an oncologist and you have your new oncologist, the very first thing they do is they say, let's walk through your clinical history. And you walk step by step by step. And so, in my record within CureWise, I happen to have done a breast cancer vaccine study at the University of Washington in 2016. And at that time, their system was not epic. And when they converted the records, it didn't get converted across. So even when I connect all my medical records together, it's always missed. It's missed by my doctor's office, et cetera. And so I added that into my curewise profile. And that's why your oncologist reviews things, and that's why we review things. You are the expert. You know your cancer, you know your history, and everything like that. So we have a chance to participate in that and share anything that's missing. And then once we have that locked and loaded, we actually have a lovely health page and it gives you your blood work. Every time I go to the oncology office and I get my blood work, I go log into Epic. I then go to test results, and then I find the test result that I care about. And then I go look at the five things that I care about with the trend chart. Now in CureWise, I just open up my application and I look at it and it shows the blood work that I care about. I can see the trend lines. And if there's anything off, I can immediately press a button and that button takes me to the chat interaction that says, Hey, this is some questions you might want to ask about your platelets or your AST or whatever blood value you want. Okay.

SPEAKER_05 16:55

It actually suggests two. Oh wow.

SPEAKER_00 16:59

It's very simple, just like you can get a trend line. It says, these are your last results. And if you're doing it at multiple facilities, we regulate because every facility calibrate their machines differently, the ranges might be different, and we calibrate and normalize all that data and we bring it all together in one spot. Then we have the chat, which is the bread and butter, which is how I use it the most. I have questions when I wake up and this is stop for a sec.

SPEAKER_05 17:22

I have two follow-up questions that I think are appropriate for the first part of our conversation. So the first one, and I remember that from our last interview, the updates that come in. We do these tests periodically, we get blood work every month, we get tumor markers, I don't know how often. So is that done automatically now? Because last time we spoke, you had to go and manually start the process of updating your records. Yeah. Still, still the case.

SPEAKER_00 17:54

No, actually, so what we've done is we currently on a weekly basis, we will go ahead and automatically refresh your medical records. But it's our best guess because your medical appointments are not shared through the CARES Act. So we don't know exactly when your medical appointments are, so we cannot automate that process to say, I had care yesterday, therefore I should be refreshed today. So we do on a routine basis update that and we will send you a notification via your email saying we've gone ahead and pulled your data and now it's been up to date. You can go take a look at it. But if you've had a particular result, you can always at any point in time choose to ask for a refresh of your data. The only caveat to that, and this is true whether you're using our system, quad, anything, any provider out there, is that it we're relying upon the medical provider to make that information available to tools like ours. And that's typically within four to 24 hours, and it's dependent upon your particular care facility. Okay. So there are times, let's say I get a scan back and I want to know immediately, I can always simply save that as a PDF and upload it directly into CureWise, and I can get immediate responses. And so there's a way as a customer, you can go ahead and always have current accurate information.

SPEAKER_05 19:16

Just a quick side note for context and clarification. What we're talking about here, it's not the pandemic stimulus bill that was known as CARES Act. It's the 21st Century Cures Act and its information blocking rule, which officially went into full effect for providers on April 5th, 2021. Before that, it was really common for patients to wait days or even pay for their own medical records, or for results to sit in a portal until the doctor reviewed them. Now, under this rule, most healthcare providers are required to give patients immediate electronic access to their electronic health information, things like lab results, medication lists, and clinical notes, unless a very narrow exception applies. If they deliberately slow that down, charge for it, or otherwise interfere with access, that can be considered information blocking, and there are real penalties on the table. For tools like your wise, that 2021 change is huge. It's what finally lets patients put their own records into a platform and start using them instead of being stuck waiting for someone else's permission. Actually, there is a question I'm gonna ask you, and Abigail is gonna kill me if I don't ask this question. She was very concerned about the security issue of data. How secure are our data? So, how do we make sure that the data that you get won't be shared with anybody else?

SPEAKER_03 20:59

So if you use online banking or if you use my chart, it's the same level of security. And understand, we're doing this for ourselves with our own medical data. We care, but we know it's high stakes, we know it's private, it's the highest grade of security, the same as banking and other medical records.

SPEAKER_05 21:16

All right, great. Linda, you're very, very polite. You raised your hand, so please go ahead and ask your question.

SPEAKER_04 21:24

My husband's in shock. Well, I just wanted to give an example of something that we found so helpful. In my chart, we're on an epic system here at Fred Hutchinson in Seattle, and that will show a trend line for tumor markers. No problem. We're monitoring bone disease in most cases. And when you're reading your pet reports, it can't show a trend line. It'll show you the measurement compared to the last interval, but it won't show you beyond that. And we have been playing around with it, and you can actually get more trend from multiple. PET reports, if you ask your wives to do that. So for trying to monitor response to therapy over the last six months, it's super helpful. And it strikes me why aren't the medical record companies able to do this? It's what they should be doing.

SPEAKER_05 22:16

It's very helpful. Todd, do you have anything to add to this example or any thoughts on the data?

SPEAKER_02 22:24

I like Lisa how you said when you come in and you set up your preferences and you can say, talk to me in medical language or something in the middle or plain language. All those kinds of settings are really good. One of the things that I think about as I did is I get concerned about listening to the internet interpretations. We've all heard this, right? If you feel sick, don't go look it up on the internet because you'll scare yourself or you'll have false hope one way or the other, it's wrong. And I think carewise has still some risk of that, but I liked somewhere in there where you had all the disclaimers saying, I'm not a doctor, this is only for observation, not interpretation. What I've done just in this experience, just to keep myself sane and also as a caregiver supporting window, is to try really hard not to do interpretations. And in fact, I went and told that I put in a rule myself and I said, Can you remember this rule, care wise, that I don't want any interpretations? I want you to warn me before you give me an interpretation or even ask my permission to give me an interpretation. I'm happy with observations all day long. Longitudinal things I didn't think of, but I don't want false hope and I don't want false fear. And so don't be my doctor. Don't give me interpretations. And it responded nicely. It said, okay, good. So I'll remember that the next time we have any chat and I have an interpretation share, I'll ask you if you want to hear it from me before I give it to you. And I'm like, okay, good. Then I've protected myself and my psyche and my partner from that. So I'm curious to get your reaction to what I've said, what you're thinking about that risk. Is that a real risk? Is that a good way to manage it? Or does CureWise have more features that I don't know about in this way?

SPEAKER_03 24:02

CureWise has pretty strict guardrails in there to absolutely not try to be your doctor, but to help educate you on things that are in your medical record and help coach you about things you might want to talk to your doctor about. But I think that the big thing that differentiates from what you were talking about at the beginning. I have some symptom or have something. I'm asking some sort of question. SecureWise is rooted in your medical record. So when you import your medical record from all these different institutions that it might be scattered around, it's analyzed that, it's created your clinical history from that, and that's what grounds everything. There's a lot of stuff in your clinical history that how would you know what it even is if you didn't go to medical school? I think that the biggest part of CureWise is hey, help explain this to me so when I'm sitting down with my doctor, I can have a better conversation so I can understand what's going on. I think that the biggest thing that's one is it's rooted in your own medical record, and two, the the guardrails around it are very strict and don't try to be anybody's doctor. That's not the job. The job is to be an educator to explain what you're seeing in the medical record. So that's the philosophy.

SPEAKER_04 25:15

I do want to reiterate that Todd had a lot of experience using just the basic AI tools before he discovered CureWise. And so maybe you had some experience around the interpretation thing with the tools that don't have the guardrails that CureWise has. And I also wanted to say that I really think that we might be skipping over sort of the most important point here. The difference between going to ChatGPT and using CureWise is this is informed by your medical record. There's a disclaimer in CureWise when you get to the point of uploading your medical record that says, hey, this might take a while, maybe an hour and a half. My experience was it was super fast and it was super easy. And I am still blown away by how easy it was, given how hard it is to get your medical records transferred for second opinions and all that kind of stuff. I also want to say that my care started in 2001 at a different institution. And I did have two institutions over the course of my first diagnosis and then my metastatic diagnosis. It was super easy to upload from both institutions. All of the data was complete and synthesized into a summary that was seamless. So I don't want to blow by that. It's just mind-blowing to me how easy you make it for the patient and family.

SPEAKER_03 26:41

Well, mine is in eight different institutions. And I was talking to my doctor yesterday about it, and she was blown away because the data, even though it's all electronic out there, it's still in these different silos, and people have a hard time putting it all together. Wait a second. We are all over the place.

SPEAKER_05 26:55

Lisa, we're going back to the chat. You started talking about the chat. So let's go there.

SPEAKER_00 27:02

You asked us earlier, what did we learn from the beta testers and things like that? And one of the things that we learned from the beta testers is that they simply wanted to be able to talk to an agent that could communicate accurate information and could understand how they wanted to be communicated with. And so while originally we let people pick, I want to talk to an oncologist or a radiation oncologist or a pathologist or those types of things. And what we landed on was something much simpler. We have a single agent called CureWise, and that agent is able to represent all of the different medical perspectives. And you can ask it, hey, I'm going to be meeting with a radiologist. What are the things that I need to care about this particular piece? Or I'm looking at either doing cyber knife for my liver spot, or I'm looking at ablation, or I'm looking at surgery. Help me understand those pieces. And so the CureWise agent has the capability to answer with the different medical perspectives and it allows you to say how you want to be communicated with. And so it's much simpler. And for most of our customers, that's just the preferred interface. And so that's why we leaned into that particular perspective.

SPEAKER_05 28:14

Let's take a quick break to explain again the word hallucination. In AI, hallucination doesn't mean seeing things, it means when an AI model confidently states something that is factually incorrect. It's one of the biggest concerns with using AI for medical information. This is exactly why CureWise builds in guardrails and cross-checks multiple models to catch those moments before they reach you. And while we're at it, another term used by CureWise may need additional explanation. It's cross-check. Cross-check is CureWise's way of getting a second and third opinion instantly. When you use it, CureWise sends your exact question along with your medical record to three different AI models simultaneously. Each one answers independently, and then a fourth model synthesizes all three responses to show you where they agree and where they disagree. It's a powerful way to make sure you're getting accurate, well-rounded information.

SPEAKER_00 29:19

And then we built another piece, which is what Steve talked about. One of the things we all worry about is hallucination. And part of hallucination is did the AI tell me the right thing? And one of the easiest ways to check that is to actually ask all the different AI models the same question. And so we have something called cross-check. And you ask your CureWise agent. I just heard about the ASCO conference and the new RAS medicine, does not apply to me. And you hit the button, and what it'll do is it'll take your medical record, and I have metastatic breast cancer and an ERBB2 meditation. I do not have pancreatic cancer, and it'll go through and ask chat, GBT, and anthropic and Gemini the same question. All of them will respond. And so you can reach each individual response. Then it goes through and says, synthesize that for me. And it'll tell you where the models are alike and where the models are different. So you can really test the waters to make sure you're getting accurate information. And as Steve mentioned earlier, the models do 95% of the time they answer the question in the same way. But there are times when they answer it slightly differently. For example, last year when I was trying to decide what I was going to do for my next treatment after I failed in hair two after seven and a half years. But Anthropic and Chat GBT both suggested that I go on to a new ADC, like looking at the ones in trial. But Gemini kept coming back to Dado DXD and what's the one that said Tuzumab, the other one that I can't remember. So it suggested both of those. And I couldn't figure out why for the longest time. And finally I asked it. And it was that Gemini was assuming that the reason why I stopped having inherited work was because the linker was broken. And that's what Gemini thought. The other two considered the payload. And when I sat down with Sarah Tulaney and my own oncologists, the real issue is that nobody knows. We don't have enough data right now to know if people are failing the linker or the payload. And so it did give me that delta by asking the different models. And so I really appreciate being able to ask all three models at once the question and get those answers. There's two features within our platform. One is normal chat where you just get to converse. And for 99% of the time, all I do, because I really only just need one answer. But there are times when I want to see what the different models say. And so then I go use a feature called cross-check. And that's when I'm able to ask a question and it allows all three models. And so all three models are given the same information. They're given the same prompt, they're given the same medical information, and they all answer the question based upon how those models have been trained.

SPEAKER_05 32:03

And based on what you're saying, we would also know which LLM models are saying what, right? You would see all three.

SPEAKER_03 32:12

So it's doing two things. One is they're each answering it independently, and then all of the responses from all the models get synthesized and it says, okay, where do they agree? Where did they disagree? Compare the results. So it's a two-step process. So it does answer independently, and then it goes through a synthesis and comparison step. So you don't have to go do that. The system is doing that for you.

SPEAKER_05 32:34

So every model gets the responses from the other two? Yeah.

SPEAKER_03 32:37

And every model No, there's a synthesizer model that gets all the responses from all three models and then it synthesizes.

SPEAKER_04 32:43

But do the models consult each other? Are you able to assess that before it reports the first level assessment? Are these models going out and incorporating their competitors?

SPEAKER_03 32:59

No, no. The models are actually so good at this point when your medical record is properly prepared and you've properly guided it. The models are so good right now in their medical training that some of the things we were doing a year ago, having the models debate each other, that's become less relevant because the models are like a hundred times better today than they were a year ago.

SPEAKER_04 33:21

I just find that answer very reassuring, Steve, the way this is all controlled. They're not commingling their data before they give you the first level report. Like that said, it really is value add here.

SPEAKER_02 33:33

But also, I'll say that this is a value add of CureWise is to take the independent answers and then cross-check them and then determine what's valuable out of that cross-check or what's relevant to the patient. I think that's a nice value add on top.

SPEAKER_03 33:47

The models are changing. Next month another one's going to come out, and a month after that is another one's going to come out. So we're doing the work to keep on top of that and make sure we're testing each one and putting the most recent best one in there. So you don't have to think about that.

SPEAKER_05 34:00

I have a question as a follow-up to the first feature where you just want to chat. You chat to some chatbot. Do you actually know who you're chatting with? Or is this something that you guys decide on the back end?

SPEAKER_03 34:13

You can choose the underlying model if you want to, but we've done a bunch of testing and we put in the as the default, the one that's been the best overall. If you decide you want to go choose a different one, you can do that.

SPEAKER_05 34:24

This is such a great tool. And the patients know that AI is here, AI is coming, and so many are worried that AI would replace the doctors. How do we assure them that's not what's happening? That this is a tool, and I wanted to mention this as a separate item, and we'll talk about this. I think it's important that your tool is also a training tool. Your tool trains the users how to talk to the doctors. But let's first talk about this. How do we assure the users that this is not going to replace the doctors, that this is a companion to you when you go see your doctor?

SPEAKER_03 35:07

I've talked to many doctors about this and many oncologists about this. And first of all, being an oncologist has got to be about the most difficult thing in the world. It's a really difficult profession to be in. And you're dealing with a lot of difficult situations. Everybody I've talked to has wanted their patients to be more educated so they can have more quality conversations, so they can be more collaborative with their patients, because so many of the decisions are made by the patient. Do you want to do this or do you want to do that? Here are the pros and cons. You decide. Now, if you have a patient that is completely uninformed about any of that, it is really difficult. It makes the job of the doctor so much more difficult. But if you have a patient that's already studied it and has learned the terminology and has already educated themselves on this, that conversation is much more collaborative. It's much more reassuring, it's a much higher quality, and it leads to better outcomes. This already has been proven that the shared decision making and collaborative care leads to better results. That is what CERWISE is doing. It's helping patients, it's helping people become educated in their disease so they can have a higher quality discussion with their oncologists so they can make better decisions together. And that's what this is for.

SPEAKER_05 36:25

So you have mentioned, and this is something I wanted to follow up on. You have mentioned that the job of an oncologist is probably one of the hardest things to do. And I just interviewed a doctor and I asked her specifically that question: what does a community general oncologist do? Unfortunately, some cancers are still not that many options, but certainly in breast cancer, there are so incredibly many options. But you keep saying, and you said that in our first interview, that this platform is not for doctors, it's for patients. Why can't a doctor use it? And what makes it so special that it is for patients, not doctors?

SPEAKER_03 37:03

So if you look at all the things going on in AI and medicine, there are a lot of people working on different things. The one part that's really been left out is something that's really just focused on serving the patient. Since we are patients, since we're doing this for ourselves, that is, of course, where we start. Because that seems to be the missing link. Every time I go to my doctor says, Hey, do you mind if I use this AI? Do you mind if I do this? A lot of people are worrying about the doctors. But we want doctors to be able to participate and be involved, but our priority is the patient. That's who we care about, that's who we serve. From the standpoint Todd talked about the caregiver interface. From next beyond the patient is everybody has somebody on their care team that's probably a family member. So that's the next priority. And then there may be, you know, down the line, there may be more professional caregivers and providers. We're going to figure that all out, but who we serve is the patient and their family. And that's always going to be our priority. So this is a great segue to talk about the training.

SPEAKER_05 38:04

How does CureWise help patients be prepared for their conversations with the doctor?

SPEAKER_03 38:13

I'll give you just an example from yesterday from myself. You know you have a very limited time with your doctor. If you just go in there quietly and you don't say anything, you're going to get very limited out of that. But if you come in prepared and you know what your questions are and you know what you want to address, and you know what you're worried about, the quality of your visit and what you get out of it is so much better. So I literally rehearsed with Kirwise for an hour because my doctor is a specialist. She's one of the top people in the country. She's really amazing. And I value that time. So I want to get every bit out of that. So I rehearsed. I talked about all my questions with Kirwise first. So when I went in, I felt like I was prepared. I was more confident. I knew what I wanted. I said, this is my number one complaint, my number one issue. And my biggest question is that not only did we have a really valuable conversation, but we actually identified that one of my other doctors made a mistake. And if I hadn't been asking the questions, we wouldn't have identified that. My doctor identified that. But it was through the digging deeper and asking questions that we got to the bottom of things. I don't think that would have happened if I didn't have a chance to practice and rehearse and get educated before that meeting. These are highly trained professionals. They have a really difficult job. They care about you, but you only have 10 minutes with them. So how do you make every minute count? It really helps to educate yourself, to practice and to figure it out as much as you can ahead of time.

SPEAKER_02 39:45

I can call on that just briefly, building on what Steve said, the preparation, but I have a little bit of what I've heard called white coat syndrome, where I sit in front of a doctor and they say something, I'll ask the first question, the answer, and then I have to process that answer. And I get locked up with overthinking it, probably. And so the way I use the cure wise a little bit was okay, if my question is this and Dr. Linden says A or B, then my next question is. And so it helped me anticipate. In addition to prepare, it also then helped me to anticipate. And again, to your point about using the time while having a rich conversation, I didn't lock up it after the first answer came. I was actually able to stay in the game and go, okay, I anticipated that. And now my next question for her is this.

SPEAKER_00 40:29

Currently, it's a self-directed experience. None of us know exactly what questions are going to happen in our medical record. Every one of them is unique. We do prompt the customer with these are some questions that you may want to ask. And just like Claude does or any of the other ones do, in our RD side of the house, I'm working with a physician's assistant. And we're thinking for a person that wants a little more guiding, we're working through maybe some flows. Hey, I have an oncology meeting. Help me prepare for it. And so you can really work through that. Is it just I'm going to walk in and I'm going to get my blood work and I'm going to walk out of the office and there really isn't any big questions? Or is it a bigger one where you have a scan result that's changed and you need to think through what I'm doing, a treatment change? And so what type of questions are you thinking about? How do you mentally work through that in advance? When should you send your oncologists the questions in advance? I've been navigating for metastatic patients for the last decade. And I always say that when you're looking at a treatment change or a scan result was bad or you're having a major new side effect, I always find it's best to send your oncologists the questions in advance so they have time to think about them. Otherwise, they're caught off guard in the meeting. And so what we will be building into the flows, guiding principles or guidance for our customers, so they have a better ability to self-navigate. But what we have right now, which is infinitely important, is having the time to ask questions and understand things so that you, whether it's white coat syndrome or just locked in fear, because we have all had that experience, right? My last scan, I got it back about 10 minutes before I saw my oncologist. And she'd mentally processed the scan. I had not. And she was moving on. And I finally had to say, Dr. Chen, we're not talking about anything till we go back to the scan, because I can't hear anything you're telling me right now because my brain is solely focused on this other piece of information. And whenever you have the time as a person to process, come to terms with something, then your mind opens up again to hear new information. And that's true for you, it's true for your caregiver and all of the people in the puzzle. And so I really love that with the CARES Act, the information is released to us immediately. So we see things. And now we have the ability with guardrails, because that is the most important thing. I need to ask questions with guardrails. I don't want some bizarre answer coming out of Chat GPT because I forgot to attach the right medical record or I didn't give it enough information or it hallucinated or did whatever. I want to know that my medical record was processed, my information was used, that the prompts that were sent made sure that I was getting information in a safe way that I can understand. It's not overstating what I should do as a patient. It's not trying to prescribe medicine. It's saying, this is what information you need, this is how I can help you understand it. You can change the metaphor, you can do all of those different pieces. And let's prepare for your oncology appointment. But your provider team is the one that prescribes the medicine, and they're such a valuable part of the team. But shared decision making makes for a really powerful relationship with your oncologist and leads to good outcomes because you feel like you have skin in the game and you're not a victim of your disease. You are an active participant navigating this journey with information and agency. Yeah. That's good.

SPEAKER_04 43:42

I've used it just recently for investigation of two clinical trials that we have on our radar and just the explanation of what is needed for these clinical trials and where there may be an eligibility issue. The explanation is so helpful. It's just far and above anything I've ever seen.

SPEAKER_05 44:04

I wanted to follow up with something that we actually touched upon earlier in our conversation. I think that this is important to mention that as cancer patients, cancer, of course, is our highest priority, and that's what we deal with, and that's predominantly what we're talking about here. But as long-time cancer patients, we develop other conditions, other diseases. Because of my cancer, I now have showgreens, autoimmune disease. And we get to see many more doctors, not just our oncologists, our radiation oncologists, and so on. And the good thing about QRYs, it's not just limited to that. You can actually ask QRYs, how can I deal with this? What should I do?

SPEAKER_00 44:50

Absolutely. So it does provide that broad context. You know, we focus on cancer because we, Steve and I, built this with our passion of how we need an agency to navigate our own cancer journey. But the underlying prompts, your medical record, it has all the information. So you can ask those broader questions. I use it right now with my dad. My dad happens to have prostate cancer and Parkinson's. I ask far more questions about Parkinson's and my father than I do about his prostate cancer because that's more urgent and present. So you can absolutely ask it broader questions than that.

SPEAKER_05 45:24

And that's really important. Lisa, we need to talk about the costs. We don't like to talk about money in general, but this is an important conversation, right? So let's talk about the costs. Why isn't it free? How much is it? I know it's subscription based. So how much is a subscription? Let's talk about that.

SPEAKER_00 45:44

Absolutely. So while we would love for the product to be free, in reality, there are real costs to pulling your medical records and allowing you to interact with the different AI models. So every time we pull your medical records, we pay a service fee for that. And every time you ask a question to an AI agent, we pay a fee for that. And currently, that's all something that we have to pay for and support out of our own revenue stream. So we need to pass that cost on to the customer. Our current model is that we will charge you either a monthly rate of $49 a month, or we offer a discounted rate of, I think it's $36 a month if you pick an annual plan. And we're absolutely looking at ways to bring those costs down because we want to be able to reach out to as many people as possible. And we're looking at ways to fund this for people that can't afford that level of resource. One of the options will be as we proceed forward, we have a lot of patients that want to get on clinical trials, and we're going to be able to potentially refer people into clinical trials. And that could be a source that we have the pharma companies represent this. We could get foundations to represent it. We want everybody to have access to this, but we don't want to hold the opportunity for people to use the service while we figure that out. So in the near term, our pricing model is going to be $49 a month for if you offer a monthly service, or we can do an annual service for $36 a month. We offer a free trial so people can see if they find value in it. And during that first seven days, we are going to ingest all of your medical records. We're going to show you that we understand you as a customer. We're going to allow you to have seven days of chatting so you can really experience it and see if it's something that works for you. We want to make sure that we're delivering value to our customers. Steve and I are doing this with the rest of the CureWise team to help patients have agency. And so we absolutely want to make sure people have access to the best possible information so they can have the best outcomes with their oncology team.

SPEAKER_05 47:40

That's wonderful. And you already have the first paying customer. Yeah. As a parting thought, what is your message to the future users of the platform?

SPEAKER_03 47:52

Steve, I guess that would be you now. My message is that the science of cancer is going really fast, and there are a lot of new things out there, and there are more and more things out there every day. You need to get educated, you need to learn about this. If you just go in and you don't ask any questions, you're probably going to get just the guidelines, which might be great, but it might be 10 years old. And there might have been a lot of progress in the last 10 years and a lot of new things coming on. And if you don't know about them, you don't ask about them, you don't learn about them, you're not going to have access to them. I know that just last year, the National Cancer Institute published a survey of this in the Journal of the NCI said that 60% of people with cancer have an actionable mutation, a druggable mutation. Only 15% are actually getting treated for it. So there's a big gap between where the science is and what most people are getting. And the only way you're going to close that gap is by learning everything you can about your disease, asking the right questions, making sure you're getting the best tests, and that you're pushing your doctors to act on that. Because the easy way out, if nobody has time, is just go to what the guidelines are. Most of the new stuff that scientists come up with in the last few years, it's beyond the guidelines. Often what they say is, well, try something beyond the guidelines only if the guidelines fail. But it's increasingly predictable about whether or not the guidelines are going to work for you. So if you know it's not going to work for you, then don't waste a year doing the wrong thing. Try to get onto something that's better matched to your biology. Everybody's cancer is different. Everybody's cancer is biologically unique. What you need to get depends on what your biology is. So you need to learn about that and advocate for it. And that's why we're here.

SPEAKER_05 49:42

Thank you. Mic drop moment.

SPEAKER_04 49:44

Yeah. So, Lisa, as we close out, what do you want people to do as soon as they close their podcast here?

SPEAKER_00 49:52

I want them to go to www.curewise.com and sign up and become one of the people that has agency and can self-advocate with their oncologists and find the best possible care for their metastatic disease.

SPEAKER_05 50:06

Excellent. And Todd, do you have anything you want to impart on our listeners before we go?

SPEAKER_02 50:14

I think it's worth a look. I think that there are many tools out there, but this adds so much value on top of the raw tools, the raw data. It brings together everything around the patient in a unique way. I don't think there's anything like it that I've seen being a caregiver and really looking forward to what's next here.

SPEAKER_05 50:31

Oh, what a wonderful way to end. And I want to say we don't get paid for this. So this is not a paid advertisement. That's why we're here. We're here only because we're excited about this product. We want to be here. We want to be the first paying customers. And good luck to you. Congratulations. And we are rooting for you. Thanks, Lisa. Thank you. Take care. And we are done. That's a wrap on today's episode of Live from Stage 4. A huge thank you to Steve Brown, Lisa Booth, Linda Weatherby, and Todd Weatherby for joining us and for sharing so openly. This is exactly the kind of conversation that makes this community what it is. In this episode, we caught up with Steve Brown and Lisa Booth, the team behind CureWise. We talked about how CureWise connects to your medical records across multiple institutions, how it helps you prepare for oncology appointments, and why having the right questions ready can literally change your treatment outcomes. We also heard from beta users Linda and Todd, a patient and a caregiver duo on what it actually feels like to use the platform for the first time. From the cross-check feature that compares multiple AI models to the Caregiver Access Roadmap, to the very real gap between available treatments and what patients are actually receiving, this one is packed. If anything you heard today resonated with you, head over to www.curewise.com to sign up. There is a free seven-day trial so you can experience it for yourself. No commitment, just information and agency in your hands where it belongs. A quick but important note: CureWise is not a doctor. It's an incredibly powerful tool to help you understand your medical records, prepare better questions, and have more informed conversations with your care team. But all medical decisions should always be made with your physician. If you enjoyed this episode, please take a moment to subscribe and leave us a review. Wherever you listen to your podcasts, it takes 30 seconds and it makes an enormous difference in helping us reach more people in this community who need to hear these conversations. Share it with someone who could use it, a fellow patient, a caregiver, a friend who just got a diagnosis and doesn't know where to start. We'll see you next time on Live From Stage 4. Take care of yourself.