Thanks, Brian. I will walk through some product updates along with the next phase of development because the numbers you just heard are downstream of it. Everything I'm about to walk through comes back to a single idea you heard from David at the open, read the patient's entire molecular story at depth, and you can transition the benefits of precision medicine from late-stage disease to early stage. And nowhere is that more beneficial than in early detection. So let's start with Caris Detect. The first thing to understand about Detect is breadth. From a single routine blood draw, Caris Detect now identifies 58 distinct cancer types, spanning solid tumors, hematological malignancies and importantly, providing guidance to patients to minimize the time to diagnostic resolution. Look across this map, lung, colorectal, breast, prostate, pancreas, the full upper GI and gynecological spectrum, skin, brain, renal, urothelial, liver and on the right, the harder categories, most screening tests simply don't touch, soft tissues and bone sarcomas and of course, the hematological malignancies. This is not a single cancer test wearing a wide label. It's a genuinely pan-cancer coverage from one blood draw, and that breadth is a direct product of building the Whole Genome and Whole Transcriptome Sequencing, which spans the entire spectrum of biology rather than a narrow approach, which only captures a small portion of what drives cancer. Detecting a signal is only half the job. The question every clinician asks next is, where is it? And where our tissue of origin classifier changes the economics of the workup. Our approach focuses on what the best action for the patient is, finding the cancer faster while minimizing the number of procedures a patient have to experience to get there. The way to think about this slide is on the left. When a signal comes back, most tests hand the physician an open-ended search, a scattershot battery of scans and procedures. Our classifier was the opposite. It concentrates probability onto the true site and turns that open-ended hunt into a short prioritized workup. Here's what that looks like in the validation data. Across true positives, a little over 2,500 patients, 83.9% are resolved in a single workup and 99.8% are localized within 2. That's an average of just 1.19 procedures per patient. And even in the small positive -- false positive group, we've resolved essentially 100% within 2 workups, meaning patients are not sent down a long, arduous, anxious and expensive diagnostic odyssey. Down at the bottom is why that matters. It's not abstract, fewer procedures, less radiation exposure, a faster path from signal to answer and real support for the hardest cases in oncology. The metastases of unknown primary where origin is genuinely uncertain and because every routing step is explainable, the coverage balance is a dial we can tune. It's not a black box. I want to make this concrete because it's ultimately what the ordering physician holds in their hands. Every detect result is delivered as a prioritized workup, not just a yes or no. At the top, a clear actionable statement, cancer signal detected backed by Whole Genome and Whole Transcriptome Sequencing. Below it, the suspected tissues of origin ranked by probability. Sometimes that's a single high confidence call like the 99% thoracic lung example on the left. Sometimes the signal is spread across sites, like the example on the right, led by HPV liver at 18%. And critically, each of those ranked sites comes with a specific next step, the exact study to order mapped on to the body, a colonoscopy, a contrast enhanced CT of the chest or a CT of the abdomen and pelvis. We even tell the physician what the data de-prioritizes, the tissues the signal makes unlikely listed at under 0.1%. So the diagnostic search has narrowed from the very first day, sometimes knowing what's -- where it's not can be just as valuable as knowing where it is. That's the Detect story. Test, know, act. Now this is the part that I'm most excited to walk you through because it's where Caris stops describing disease and starts intervening against it. We call it the mutational cleanse, and it's the embodiment of that shift you saw on the title from personalized medicine to personalized prevention. Here's the arc, left to right. First, Caris Detects flags disease early, while tissue of origin routing is possible and disease burden is still low. In the validation set, that's 60% Stage 1/II sensitivity or catching, and this is the stage that has a very high cure rate. We don't just want to report the signal. We want to follow it. Step 2, Caris MAX. We go back and interrogate the circulating tumor signal at 10,000x depth of coverage, ultra-deep mutational analysis of the exome layered with HLA and germline logic to separate out real somatic mutations from noise and identify the rare variants that matter. Step 3 is where our AI does the work no panel can. It scores each candidate mutation on pathogenicity, clonality, expression, antigen processing, HLA fit and blood on-target risk to identify the subset of mutations that are immunogenic, which are the mutations the immune system can actually see. And the performance here is strong, 83.8% positive predictive value and 86.5% sensitivity on the top variant per patient. We were able to achieve this level of performance by leveraging our unmatched data set, which contains thousands of specimens collected before the administration of immunotherapy and matched samples after. Step 4, the top neoepitopes become patient-specific immune targets, which we can monitor over time against ctDNA and T-cell response to determine if the source of the signal is going away, a closed loop. So the whole idea in one line is on this slide, find the dangerous clone early, make its mutation visible to the immune system and remove it before clinically overt disease ever emerges. That is early detection becoming early interception. Now let me turn from the frontier to what's landing in the near term, 3 pipeline items an upgrade to MI Clarity, our MRD program and the clinical evidence underneath all of it. MI Clarity is our recurrence risk platform, and version 2 meaningfully expands what it does. Version 1 already delivered distant recurrence across both the early and late windows, years 0 through 5 and 5 through 15, orderable right at diagnosis with fast turnaround time at an accessible cost. What's new in V2 is decision support, not just prognosis. We're adding chemotherapy decision support, identifying which patients are actually likely to benefit from chemo. We have also extended endocrine therapy decision support, informing treatment beyond the first 5 years. And we have also expanded ordering years after diagnosis, which the extended endocrine therapy decisions need to be made. And finally, we've integrated early and late treatment decision support into a single test. In short, version 2 moves MI Clarity from telling you the risk to helping you act on it. As you heard at the open, we're taking the platform into MRD, minimal residual disease, and we're doing it with 2 complementary approaches because different clinical settings need different tools. On the left, tumor-naive built on Whole Exome plus Whole Transcriptome on our Caris Assure platform initially in colorectal, a diagnostic for Stage 2 and 3 solid tumors after curative intent treatment, profiling cancer-associated circulating tumor DNA and RNA from a whole blood sample with no need for the original tumor tissue. We're collecting more longitudinal outcome data for MolDX technical assessment and with more indications to follow. On the right, a tumor-informed Whole Genome solution, leveraging our Caris Precision technology platform, we perform Whole Genome not just on the tissue, but also on the blood sample. In the approach, we will offer pan-tumor Stage 1 through 3. This comprehensive approach reflects our forward-looking vision of always providing the best possible assays. Tumor normal Whole Genome Sequencing identifies the maximum number of trackers, which minimizes false negatives and drives ultra-low parts per million sensitivity. The analytic performance speaks for itself over 5 logs of linear dynamic range, a median of roughly 15,000 trackers per patient and an R squared above 0.99 with a slope near 1 across that entire range. The validation for this assay is in process and our launch planning is underway. We are applying the same principle as we always use everywhere else on the Caris platform, more depth, more trackers and fewer things missed. And I want to close on the studies we released this quarter to further build on the evidence that our approach leads to better outcomes for patients compared to small panels of hundreds of genes because this is the through line of the entire company. Comprehensive testing reveals what targeted gene panels miss on both sides of the equation, who's eligible for therapy and how they actually do. Two peer-reviewed studies from this year support this claim. On the left, our lookback program published in the oncologist shows how our commitment to the patient doesn't end when we deliver the report. By re-interrogating prior comprehensive results with no new test and no re-biopsy, we identified 13,293 patients newly eligible for FDA-approved targeted therapies and told their physicians about their new options. That came from reviewing 87 FDA approvals across more than 483,000 molecular profiles in 10 tumor types. The depth we captured years ago is still generating new treatment options for patients still fighting their disease today. On the right, published in cancer immunology, the study shows that our whole exome-based total mutational burden assay drove longer overall survival than smaller panels when selecting patients for pembrolizumab. And you can see why. Targeted panels, even larger ones at 300 to 650 genes, disagreed with the Whole Exome on TMB in roughly 10% to 15% of cases. That's 1 in 7 to 1 in 10 patients potentially misscored on a decision determines whether they get immunotherapy and the opportunity to live longer. And we don't just view this as a competitive advantage. We do it because it's providing patients with the best care, and it's why we built the entire platform the way we have. So that's our focus on our science, detection that's both broad and precise, an intersection strategy that's genuinely novel and world-changing and a pipeline landing in the near term, all on evidence that keeps validating the depth-first approach. With that, I'll turn it over to Luke.