{
  "docId": "019dd923-622c-750b-8b98-0c85e40ec39b",
  "docSlug": "9b248b07037e",
  "documentTitle": "CareDx, Inc. (CDNA)",
  "authorId": "53_Kerrisdale_Capital",
  "authorName": "Kerrisdale Capital",
  "documentKindSlug": "research-note",
  "documentKindLabel": "Research note",
  "sourceTypeSlug": "short_seller",
  "sourceTypeLabel": "Short seller",
  "presentationDate": "2019-07-01 00:00:00",
  "orientation": "portrait",
  "aspectRatio": 0.77272725,
  "pageNumber": 13,
  "pageCount": 29,
  "prevPage": 12,
  "nextPage": 14,
  "slideType": "expose_contradiction",
  "function": "expose_contradiction",
  "density": "dense",
  "nDataPoints": 6,
  "notes": "This is a research note from Kerrisdale Capital critiquing a medical diagnostic product.",
  "elementsJson": [
    "paragraph",
    "bullet_list",
    "footnote"
  ],
  "metadataConfidence": 1,
  "imagePath": null,
  "slideHref": "/slides/019dd923-622c-750b-8b98-0c85e40ec39b/13",
  "deckHref": "/decks/019dd923-622c-750b-8b98-0c85e40ec39b",
  "deckJsonHref": "/decks/019dd923-622c-750b-8b98-0c85e40ec39b.json",
  "deckAnchorHref": "/decks/019dd923-622c-750b-8b98-0c85e40ec39b#slide-13",
  "components": [
    {
      "bbox": null,
      "kind": "callout",
      "text": "But missing 12 cases of rejection out of 27 is an unmitigated failure, and reveals the inability – and risk – of using AlloSure to rule out rejection.",
      "attrs": null,
      "subkind": null,
      "toolName": "Visual emphasis",
      "toolSlug": "visual-emphasis",
      "confidence": null,
      "componentId": "019dd953-203e-77ad-a69d-f61b46018fd6",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.07,
        "w": 0.78,
        "x": 0.11,
        "y": 0.34
      },
      "kind": "list",
      "text": "It’s more accurate than creatinine in discriminating rejection.\nIt can therefore be used to rule out rejection if creatinine is positive but AlloSure is negative.",
      "attrs": null,
      "subkind": "bullet",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "920a6d06-8637-4429-95ed-8e95586eca77",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": null,
      "kind": "metric",
      "text": "sensitivity: 41%",
      "attrs": null,
      "subkind": "primary",
      "toolName": "Quantification",
      "toolSlug": "quantification",
      "confidence": null,
      "componentId": "019dd953-203e-77ad-a69d-fa2a27bd5dde",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.22,
        "w": 0.78,
        "x": 0.11,
        "y": 0.12
      },
      "kind": "paragraph",
      "text": "Even with a biased sample (every sample was from a patient with some sort of kidney dysfunction as measured by creatinine/proteinuria), the data is damning. “Sensitivity” is a diagnostic test’s ability to correctly detect patients who have the condition being tested for. Using 1% as the diagnostic threshold,17 dd-cfDNA had a sensitivity of 59%. In other words, in the biopsies diagnosed as rejection, AlloSure would have detected only 59% of them, and would have missed 41% of rejection episodes. Can AlloSure be used to avoid unnecessary biopsies? Maybe, but apparently at the cost of missing 40% of the necessary biopsies.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "1eb2165d-9b3a-4a0c-aabe-6e07d80a2c6e",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.3,
        "w": 0.78,
        "x": 0.11,
        "y": 0.42
      },
      "kind": "paragraph",
      "text": "It’s technically true that, in this particular study, AlloSure was more accurate than creatinine. But “accuracy” is a statistical term, an artifact of the number of false positives and false negatives. Accuracy doesn’t account for the severity of the false result. In the case of kidney rejection, a false negative can result in organ failure, while a false positive means a relatively low-risk, but unnecessary biopsy. You’d rather over-diagnose than under-diagnose given the consequences. Creatinine indeed chalked up lots of false positives in the study (39%), while AlloSure had a lower rate of false positives (39%) and a seemingly low rate of false negatives (16%). CareDx touts the low false negative rate to show that AlloSure can be used to rule out rejection. But that’s incredibly misleading: it’s true that of 77 AlloSure tests that came back negative, “only” 12 tested positive for rejection on biopsy (16%). But missing 12 cases of rejection out of 27 is an unmitigated failure, and reveals the inability – and risk – of using AlloSure to rule out rejection.",
      "attrs": null,
      "subkind": "paragraph",
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "6052d3ea-31ad-4028-bc8b-0ff6c10390d8",
      "frameworkName": null,
      "frameworkSlug": null
    },
    {
      "bbox": {
        "h": 0.08,
        "w": 0.78,
        "x": 0.11,
        "y": 0.85
      },
      "kind": "source-note",
      "text": "17 The 1% threshold is defined in terms of the proportion of cfDNA in the blood that can be identified as donor-derived. An AlloSure >1% comes back positive, and <1% it comes back negative. This is another significant bias of the study as the threshold was chosen by the authors on an ex-post basis, and conveniently optimizes for statistical accuracy.",
      "attrs": null,
      "subkind": null,
      "toolName": null,
      "toolSlug": null,
      "confidence": null,
      "componentId": "b04d7e8b-f8dc-4e4a-82a9-7324ac162147",
      "frameworkName": null,
      "frameworkSlug": null
    }
  ],
  "metrics": [],
  "tools": [],
  "frameworks": [
    {
      "name": "fraud-exposure",
      "slug": null,
      "matchId": "579d1e70-3da0-4ee8-8246-4fa2076fcd9e",
      "evidence": "The slide systematically deconstructs the statistical claims of a medical test to expose misleading marketing.",
      "confidence": 0.9
    }
  ],
  "arcBeats": [
    {
      "to": 16,
      "from": 12,
      "beatId": "98a96039-52a0-4f39-824a-abc50d3c23dc",
      "arcName": "Problem-Agitate-Solution",
      "arcSlug": "problem-agitate-solution",
      "beatName": "Villain Critique",
      "beatSlug": null,
      "evidence": "Slides 12-16 critique AlloSure's clinical data and utility",
      "position": 1,
      "confidence": 0.8,
      "parentBeatName": null,
      "parentBeatSlug": null
    },
    {
      "to": 14,
      "from": 13,
      "beatId": "85715bb9-0467-4cac-b70b-dcb33d24e2ba",
      "arcName": "Problem-Agitate-Solution",
      "arcSlug": "problem-agitate-solution",
      "beatName": "Expose Contradiction",
      "beatSlug": null,
      "evidence": "Slides 13-14 expose contradictions in AlloSure's clinical data",
      "position": 2,
      "confidence": 0.8,
      "parentBeatName": null,
      "parentBeatSlug": null
    }
  ],
  "loops": [
    {
      "to": 26,
      "from": 4,
      "name": "Cost Of Inaction",
      "slug": "27-cost-of-inaction",
      "bestFor": "Urgent budget requests, compliance, risk mitigation",
      "matchId": "6ece1bb9-6dc2-4b36-a983-1e6b3a2a754d",
      "evidence": "The document emphasizes the risks of not acting on AlloSure's ineffectiveness",
      "position": 0,
      "objective": "Highlight the cost of inaction for investors",
      "structure": "The Status Quo -> The Hidden Costs Accumulating -> The Future State of Inaction -> The Tipping Point",
      "confidence": 0.7,
      "description": "Quantify what happens if the audience does nothing"
    }
  ],
  "imagePathAlt": null,
  "thumbSrc": null,
  "thumbSrcAlt": null,
  "locked": true
}