idea-012 · narrative
A synthesis across this candidate's claims, dossier and decks — every statement here traces to something that already exists elsewhere in the knowledge base. A narrative never invents a fact, upgrades a tag, or recommends anything.
How to read this. This is a synthesis, not a new source of truth: every claim it cites carries the tag the candidate file gives it right now, copied not paraphrased, and every line quoted from a deck is quoted verbatim. See docs/narrative-spec.md.
This is a veterinary-only continuous glucose monitor for diabetic dogs and cats — an amperometric glucose-oxidase sensor, re-engineered from decades-old human CGM sensing chemistry into a haired-skin adhesive form factor with a species-specific calibration algorithm, paired with a reader, meant to replace serial blood-glucose-curve venipuncture sampling for insulin-dose titration and hypoglycemia detection. Because FDA does not premarket-review veterinary-only devices, the intended use puts this candidate entirely outside the 510(k)/PMA system — the one settled fact in the file [claim 1: verified] — and because no veterinary predicate can exist by design, every clinical and behavioral comparison in the file is to a human device used off-label, chiefly Abbott's FreeStyle Libre. Almost everything downstream of that regulatory floor is open: no current US-specific population figure for diagnosed diabetic dogs and cats exists, no price or reachable-share figure exists, a real second-pass search for pet-insurance device-coverage precedent came back negative, and this candidate's two central engineering claims — a species-specific adhesive meant to beat the off-label analog's high failure rate, and a calibration algorithm meant to correct a documented low-glucose bias — are both unverified assumptions the decks surfaced rather than tested results. All three decks were built 2026-08-30, current against this candidate as it now stands.
The device is built on amperometric glucose-oxidase sensing chemistry — the same class underlying first-generation human CGMs — re-engineered into a haired-skin adhesive form factor with a species-specific calibration algorithm, for insulin-dose titration and hypoglycemia detection in diabetic dogs and cats. Its regulatory shape is unusual rather than merely narrow: because FDA does not premarket-review veterinary-only devices, the intended use puts this candidate outside the entire 510(k)/PMA system, a conclusion independently confirmed across four CFR sections rather than assumed [claim 1: verified]. The frontmatter also cites 21 CFR 862.1355 as a human-use CGM counterfactual under the name "Glucose test system" — that name in fact belongs to the adjacent 862.1345, and while 862.1355 is confirmed real, Class II, and does cover human CGM products under its own name, the specific section-to-name pairing this claim states is refuted [claim 2: refuted]; it does not touch the regulatory-scope conclusion, which rests entirely on claim 1. No veterinary predicate exists or can exist by design — the only analogs are human CGMs used off-label, chiefly Abbott's FreeStyle Libre, and whether that off-label use and its published accuracy literature is as described is itself unconfirmed [claim 3: unconfirmed].
The underlying chemistry's own history is unconfirmed rather than settled: it is attributed to Clark and Lyons' 1962 enzyme-electrode concept and Updike and Hicks' 1967 practical immobilization step [claim 5: unconfirmed], but the one review retrieved at abstract depth does not itself state that specific attribution, only that it is subject-consistent [claim 16: unconfirmed]. Freedom to operate is a live, unresolved gap rather than a cleared question in either direction: no specific patent has been identified to search against, and the one connector probe attempted returns unconfirmed unconditionally because PATENTSVIEW_API_KEY is not set in this environment [claim 6: unconfirmed].
On population size, the best US-specific figures in the file are dated hospital-presentation prevalence trends from 1970-1999 — rising from 19 to 64 cases per 10,000 admissions/year in dogs [claim 17: unconfirmed] and from 8 to 124 per 10,000 in cats [claim 18: unconfirmed] — and the general claim that diabetes is a common endocrine disease in dogs and cats carries no sourced figure at all [claim 7: unconfirmed]. Two research rounds since then have added larger, more recent comparators — UK primary-care feline prevalence of 0.39% in 2019 [claim 24: unconfirmed] and Japanese pet-insurance canine prevalence rising from 0.07% to 0.30% between 2015 and 2023 [claim 25: unconfirmed] — but both are explicitly non-US, and a dedicated, repeatedly-broadened search for a current US-specific figure found none: the domain dossier records this as the single largest gap standing behind the TAM's population component, closer to a settled absence than an unsearched one [dossier: veterinary-diabetes-glucose-monitoring]. The TAM's other two components — per-unit price and reachable share — are equally unestablished: no systematic pricing survey exists, only a single veterinary clinic's posted rate for the off-label analog, $75.08 per 14-day sensor [claim 19: unconfirmed]; and owner/clinician willingness to pay a premium out of pocket is stated in the file as an explicit design assumption, not a fact, since the adoption connector returns unverifiable for this class of question by design [claim 9: unconfirmed]. All three components of the TAM decomposition stay unconfirmed together [claim 8: unconfirmed].
Reimbursement runs the same way: because a veterinary-only device carries no CPT/HCPCS code or federal coverage policy, any payment would have to come through pet insurance, and no connector in this repo reaches that data [claim 10: unconfirmed]. The domain dossier records a real, targeted second-pass search for a citable pet-insurance device-coverage source — including a full-text read of one published insurance-uptake study that turned out to be non-US and silent on devices, and repeated bot-gated attempts to reach NAPHIA's industry report — and calls the result a settled negative, not an unsearched gap [dossier: veterinary-diabetes-glucose-monitoring].
The clinical-accuracy picture for the off-label analog is where the most literature sits, and it cuts both ways. Published studies report the class is generally reliable at normal and high glucose but more variable at low glucose [claim 4: unconfirmed]; a feline clamp study found the sensor underestimates glucose in euglycemia and mild hypoglycemia but overestimates it in marked hypoglycemia below 55 mg/dL — the direction of bias that matters most for hypoglycemia detection [claim 12: unconfirmed] — while a separate feline field study reported a MARD of 13.6% with 99.7% of readings in Parkes zones A+B [claim 11: unconfirmed]. Sensor durability is a documented weak point: 70% of sensors stopped functioning before the labeled 14-day wear period in one feline cohort, with a median functional life of 8.3 days [claim 13: unconfirmed]. Canine accuracy is skin-thickness dependent, reliable only above roughly 5mm of skin thickness [claim 14: unconfirmed], and a separate canine hypoglycemia study found limited analytical accuracy (50.9% vs. 67.2% for a portable meter within ±15 mg/dL) that nonetheless rarely produced clinically dangerous readings (88.6% in Parkes zones A+B) [claim 15: unconfirmed]. openFDA records recalls on the identical physical device used off-label — erroneously high glucose readings risking missed hypoglycemia [claim 21: unconfirmed], a manufacturing carbon-buildup issue risking erroneously low readings [claim 22: unconfirmed], and a battery fire-hazard recall [claim 23: unconfirmed].
The closest evidence on demand is a revealed-preference twin, not a test of this candidate's own product: in a 2023 survey of 50 diabetic-pet owners already using the off-label FreeStyle Libre, more than 80% rated it easier and less stressful than blood-glucose-curve sampling and 92% reported better diabetes control since switching, while the same owners named sensor fixation, premature detachment, and long-term affordability as their top frictions [claim 20: unconfirmed]. The domain dossier frames the in-hospital alternative's own burden the same way — stress-induced hyperglycemia is a measured, not assumed, confound on the clinic glucose curve [dossier: veterinary-diabetes-glucose-monitoring] — which is context for why a home-readable device has appeal independent of any comfort claim, not evidence that owners would adopt this not-yet-built device specifically. Every engineering and behavioral condition specific to this candidate's own product remains an unverified write-back the three decks put on record rather than a tested result: a species-specific adhesive beating the off-label analog's 70% failure rate [claim 26: unverified]; a veterinarian trusting the device for hypoglycemia detection despite the documented low-glucose bias [claim 27: unverified]; whether ALR Technologies' GluCurve Pet CGM — the one purpose-built veterinary competitor identified in the domain dossier, whose 2023-to-2026 US availability gap and absence of any published accuracy or complication literature the dossier records but no claim states — is already commercially competitive [claim 28: unverified] [dossier: veterinary-diabetes-glucose-monitoring]; whether the absence of FDA premarket review materially reduces development cost and time-to-revenue [claim 29: unverified]; whether the candidate's own calibration algorithm corrects the documented biases [claim 30: unverified]; and whether the adhesive and sensor materials are biocompatible for repeated haired-skin contact [claim 31: unverified].
The desirability deck (built 2026-08-30) turns on the sharpest version of "who else has to say yes": whether a veterinarian would actually trust this device for hypoglycemia detection, given that the intended use names hypoglycemia detection explicitly and the off-label analog's documented bias runs the wrong way at exactly the range that matters — overestimating blood glucose in marked hypoglycemia rather than underestimating it [claim 12: unconfirmed], [claim 27: unverified]. Whether the candidate's own calibration algorithm corrects that bias is a question the feasibility deck carries as its own condition [claim 30: unverified] — one place two lenses converge on the same unresolved mechanism from different chairs.
"whether a veterinarian would rely on this device for hypoglycemia detection, given that the off-label analog's own underlying sensing technology is documented to overestimate blood glucose during marked hypoglycemia (<55 mg/dL) in cats" — the load-bearing condition of the desirability deck (built 2026-08-30).
The viability deck (built 2026-08-30) turns on the TAM decomposition itself — diagnosed population, price, and reachable share — because every other viability question (channel, capital, moat, competitive position) is moot without a paying population underneath it, and TAM below floor is the one rubric condition this deck treats as an automatic kill if it ever resolved that way; it is currently blank, not below floor, but nothing in the file rules a floor-level outcome out [claim 8: unconfirmed]. This is the same population gap the desirability deck and the domain dossier both name from the epidemiology side — no current US-specific figure exists anywhere in the file [claim 7: unconfirmed], [claim 17: unconfirmed], [claim 18: unconfirmed].
"the TAM decomposition — diagnosed population x price x reachable share" — the load-bearing condition of the viability deck (built 2026-08-30).
The feasibility deck (built 2026-08-30) turns on freedom to operate — the one condition across all three decks with an automatic-kill property under the Stage 5 gate — and names it as open for a tooling reason rather than a genuine unknown: the one connector call that could move it is blocked by a missing PATENTSVIEW_API_KEY, not by an unresolvable question [claim 6: unconfirmed]. That distinguishes it from this deck's other central engineering condition, whether the calibration algorithm corrects the documented low-glucose bias [claim 30: unverified], which has no comparable shortcut — it needs an actual bench or clamp study, not a credential fix.
"freedom to operate on the sensing chemistry, the adhesive/insertion mechanism, and the calibration algorithm" — the load-bearing condition of the feasibility deck (built 2026-08-30).
With freedom to operate currently blocked by a missing credential rather than a resolved finding, and no current US-specific diagnosed-population figure anywhere in the file, does either gap — once closed — actually change the answer to the desirability deck's own question: would a veterinarian trust this device for hypoglycemia detection, given that the off-label analog's documented bias runs the wrong way at exactly the glucose range where a missed low reading matters most, and this candidate's own calibration algorithm has never been tested against that bias?