evipedia.ai
← Back

Kylo-11 for Health & Longevity

Evidence Review created on 08/29/2026 using AI4L / Opus 5

Motivation

Kylo-11 is an experimental injectable drug designed to switch off the liver’s production of lipoprotein(a), a cholesterol-carrying particle in the blood whose level is set almost entirely by inherited genes. Unlike most blood fats, lipoprotein(a) barely moves with diet, exercise, or weight loss, and no approved medicine lowers it as its main purpose.

Roughly one in five people worldwide carries a high level, and that level tracks with earlier narrowing of the arteries and with stiffening of the heart’s aortic valve. For decades this was a number a person could measure but not change. A wave of gene-silencing injections has changed that, and Kylo-11 belongs to this wave; its distinguishing claim is that a single injection may hold the level down for about a year.

This review examines what is known about Kylo-11: how it works, what its early human testing has shown, what remains untested, what risks are plausible, and how it compares with the other gene-silencing drugs aimed at the same target.

Benefits - Risks - Protocol - Conclusion

This section collects high-level, non-systematic material on lipoprotein(a) — abbreviated Lp(a) (a cholesterol-carrying blood particle whose concentration is largely inherited) — and on the gene-silencing RNA (ribonucleic acid) drugs, Kylo-11 among them, built to lower it.

Note on the priority platforms: of the six, only Peter Attia and Chris Kresser carry substantial dedicated material. Searches of foundmyfitness.com, hubermanlab.com, lifeextension.com, and lifespan.io returned no dedicated treatment of Kylo-11 or of Lp(a)-lowering RNA therapeutics, so the remaining three slots were filled from primary trial reports and a journal commentary rather than padded.

Grokipedia

No Grokipedia article on Kylo-11 exists.

Examine

No Examine article on Kylo-11 exists. Examine.com covers dietary supplements and nutrition compounds; it does not cover investigational prescription-only drugs, which is the category Kylo-11 falls into.

ConsumerLab

No ConsumerLab article on Kylo-11 exists. ConsumerLab performs independent purity and potency testing of commercially sold supplements; it does not review investigational prescription-only drugs such as Kylo-11.

Systematic Reviews

No systematic review or meta-analysis addresses Kylo-11 itself, so the papers below cover its drug class — small interfering RNA (siRNA, short RNA molecules that destroy a chosen genetic message) agents that lower lipoprotein(a) to prevent atherosclerotic cardiovascular disease (ASCVD, the artery-clogging process behind most heart attacks and strokes) — on both the efficacy and the safety side.

Both sides of the trade-off are represented above: papers one and two on the claimed effect, papers three and four on adverse events, and paper five on what is given up by staying with conventional therapy. No systematic review or meta-analysis of hard cardiovascular outcomes after Lp(a) lowering exists, because no outcome trial in the class has reported.

Mechanism of Action

Kylo-11 is a small interfering RNA. Its guide strand is loaded into the RNA-induced silencing complex (RISC, the enzyme assembly that cuts genetic messages), which then cleaves LPA messenger RNA inside liver cells. Less apolipoprotein(a) is manufactured, so fewer Lp(a) particles can be assembled and released.

Delivery is the design problem. Like every clinical Lp(a) siRNA, Kylo-11 is joined to a sugar ligand recognized by the asialoglycoprotein receptor (ASGPR, an uptake receptor found almost only on liver cells). Selectivity therefore has two layers: receptor-restricted entry and sequence-specific cutting. Beyond that the chemistry is undisclosed, and every published detail traces to the developer — Kylonova / Hygieia, since 2026 part of Sino Biopharmaceutical — a party with a direct financial interest in the compound’s adoption.

Pharmacologically the class behaves unusually. Plasma half-life is hours, because unbound drug is destroyed by nucleases (enzymes that chop up nucleic acids) and cleared by the kidney; the half-life of the effect is months, because it is set by how long the loaded complex persists inside the hepatocyte (liver cell). There is no cytochrome P450 (CYP, the liver’s main drug-metabolizing enzyme family) metabolism and no meaningful transporter handling.

Two mechanistic readings compete. One holds that Lp(a) harms mainly through the cholesterol and oxidized phospholipids (fat molecules damaged by oxygen) it carries, so removal should be protective; the other holds that apolipoprotein(a)’s resemblance to plasminogen (the protein that starts clot breakdown) drives harm through clotting pathways, in which case benefit depends on completeness of clearance.

Historical Context & Evolution

Kylo-11 was not repurposed from another use. It was built for one job, and that job only became conceivable once three older lines of work converged.

The first was descriptive. Lipoprotein(a) was identified in 1963 as an inherited serum variant, and in 1987 its distinguishing protein was cloned and found to be a near-copy of plasminogen. That structural surprise installed the two competing intuitions — a lipid particle, or a clotting-system impostor — that the field still argues over.

The second was epidemiological. Pooled cohort data established a continuous association with coronary events, and Mendelian randomization (using inherited gene variants as a natural experiment to test cause) removed most of the confounding objection by showing that people born with LPA variants raising Lp(a) suffer more myocardial infarction (heart attack). That is why Lp(a) came to be treated as a target rather than a marker.

The third was pharmacological failure. Niacin, hormone therapy, and apheresis (filtering the particle out of the blood) could move Lp(a) but were non-specific, poorly tolerated, or impossible to scale, and statins do not lower it at all. Antisense drugs — short synthetic strands that block a genetic message — arrived in the 2010s and siRNA in the 2020s, after which the competition shifted from potency to dosing interval. Kylo-11 entered first-in-human testing in 2024 explicitly on that axis.

Expected Benefits

High 🟩 🟩 🟩

No benefit reaches High: every human observation of Kylo-11 comes from a single first-in-human dose-ranging trial, so no outcome of any class has been reproduced in a second trial of this compound.

Medium 🟩 🟩

Deep, Sustained Lowering of Lipoprotein(a) from a Single Injection

A single subcutaneous dose lowered Lp(a) by more than 90% at doses of 225 mg and above, with the reduction still present at 48 weeks in its first-in-human trial. The evidence is one trial in 71 people aged 18–55, dose-ranged from 9 mg to 600 mg at a single site, and funded by the sponsor, a party with a direct financial interest in the result. Independent agents silencing the same message reproduce comparable depth and duration (Nissen et al.; O’Donoghue et al.).

Magnitude: Median reduction in serum Lp(a) at 48 weeks rose with dose, from roughly 53% at 9 mg to roughly 97% at 600 mg; in the cohort entering above 200 nmol/L, 225 mg gave roughly 96%, an absolute fall of about 208 nmol/L.

Low 🟩

Reduction in Apolipoprotein B and in Measured Low-Density Lipoprotein Cholesterol

Part of what a standard panel reports as LDL (low-density lipoprotein) cholesterol rides on Lp(a) particles, so deep Lp(a) lowering also drops measured apolipoprotein B and LDL cholesterol. Quantified for a same-target antisense agent (Yeang et al.), not for Kylo-11. Partly an accounting effect, not fewer artery-damaging particles.

Magnitude: Direction is a small decrease, holding wherever baseline Lp(a) is high enough to contribute measurably to the panel; corrected-LDL analysis of the same-target antisense agent shows most of the apparent fall is Lp(a)-cholesterol being removed (Yeang et al.), and the literature reports no outcome figure for Kylo-11.

Speculative 🟨

Reduction in Atherosclerotic Cardiovascular Events

No Lp(a)-lowering drug has yet reported a cardiovascular outcome trial. The basis is therefore genetic inference alone: people born with lifelong lower Lp(a) suffer fewer heart attacks.

Slowing of Calcific Aortic Valve Disease

Genetic variants raising Lp(a) also raise aortic valve calcification risk. No trial of any Lp(a)-lowering agent has measured valve endpoints, so the basis here is mechanistic and genetic only.

Lowering of the Oxidized Phospholipid Burden Carried by Lipoprotein(a)

Lp(a) carries oxidized phospholipids, thought to drive plaque inflammation. Silencing lowers that burden for a same-class agent in a trial substudy, but the marker is unvalidated against outcomes and untested for Kylo-11.

Benefit-Modifying Factors

  • Baseline Lp(a) concentration: A percentage reduction delivers far more absolute change from 250 nmol/L than from 80 nmol/L. Trial entry required at least 75 nmol/L, and one cohort was restricted to above 200 nmol/L, so the tested benefit is a high-baseline benefit.

  • LPA kringle IV type-2 repeat number: The number of repeated segments in the LPA gene sets both particle size and how much Lp(a) is made. Fewer repeats mean small particles, high concentrations, and the largest absolute gain from silencing.

  • Sex and menopausal status: Lp(a) rises across the menopausal transition as estrogen falls, so a woman’s pre-treatment level may be a moving target. No sex-stratified efficacy data exist for Kylo-11; the first-in-human trial required participants to be non-childbearing.

  • Established atherosclerotic disease: Absolute benefit scales with underlying event risk. The phase 2 program deliberately enrolled participants with established ASCVD rather than healthy volunteers, because that is where any downstream advantage would be largest.

  • Age and cumulative exposure: Genetic evidence describes lifelong exposure, whereas treatment starts late. An adult beginning in their sixties has already accumulated decades of arterial exposure, so the recoverable share of the genetically inferred benefit is smaller than for someone starting at forty.

Potential Risks & Side Effects

High 🟥 🟥 🟥

No risk reaches High: the only human safety data on Kylo-11 are adverse-event counts from one first-in-human trial, so no adverse effect has been documented for this compound in more than one trial.

Medium 🟥 🟥

An Effect That Cannot Be Withdrawn for Roughly a Year

The property that makes annual dosing attractive also removes the option of stopping. If a delayed adverse effect emerged, or if Lp(a) needed to be restored, there is no antidote and no way to accelerate recovery; concentrations return only as liver cells clear the loaded complex. Off-treatment data for a same-class agent show Lp(a) still well below baseline a year after the last dose (O’Donoghue et al.). For Kylo-11 the twelve-month claim is simultaneously the efficacy story and this risk.

Magnitude: The figure is the duration of unremovable exposure — roughly 48 weeks of continued suppression after one dose in the Kylo-11 trial, with same-class off-treatment data showing incomplete recovery at one year (O’Donoghue et al.).

Low 🟥

Injection-Site Reactions

Redness, pain, or swelling where the drug enters the skin is the characteristic adverse event of subcutaneous sugar-conjugated gene-silencing drugs. Kylo-11’s own first-in-human trial recorded none (Sarraju et al.), so the expectation rests entirely on other agents in the class (Ray et al.; Nissen et al.).

Magnitude: In the approved sugar-conjugated gene-silencing comparator, injection-site reactions occurred in 2.6% and 4.7% of treated participants across its two phase 3 trials, against 0.9% and 0.5% on placebo, and were generally mild (Ray et al.); none at all occurred in Kylo-11’s own trial.

Transient Liver Enzyme Elevation

Hepatocyte-directed delivery makes mild rises in ALT and AST (alanine and aspartate aminotransferase, enzymes that leak when liver cells are stressed) the expected signal. Class trials report them as uncommon and self-limiting (Nissen et al.); Kylo-11’s trial reported no drug-related adverse events and no liver signal (Sarraju et al.).

Magnitude: Not quantified in available studies. Neither Kylo-11’s published trial nor the class trials give a transaminase-elevation rate; both report the finding as the absence of a signal.

Misreading of a Routine Lipid Panel After Treatment

Removing Lp(a) lowers measured LDL cholesterol without changing the artery-damaging particle count that matters most, so a post-treatment panel can look better than the underlying risk warrants and invite an inappropriate reduction of other lipid therapy (Yeang et al.). The evidence is indirect and comes from a different agent.

Magnitude: Direction is a falsely reassuring drop in measured LDL cholesterol, holding wherever baseline Lp(a) is high; corrected-LDL analysis shows most of the fall is Lp(a)-carried cholesterol (Yeang et al.), and the literature reports no figure for how often this changes a treatment decision.

Speculative 🟨

Consequences of Near-Total, Long-Term Suppression of Lipoprotein(a)

Apolipoprotein(a) resembles plasminogen and may retain roles in wound repair and innate immunity. People with genetically very low Lp(a) appear healthy, but nobody has been suppressed pharmacologically for decades.

Metabolic Signal Attached to Low Lipoprotein(a)

Low Lp(a) tracks with type 2 diabetes, but a Mendelian randomization study concluded concentration alone is probably not causal; the signal followed particle size (Kamstrup & Nordestgaard). Pharmacological lowering remains untested against this endpoint.

Risk-Modifying Factors

  • LPA isoform genotype: Small-isoform carriers have the highest baseline Lp(a) and the deepest absolute suppression, so any risk tied to near-total removal of the particle would surface in them first.

  • Baseline liver panel: A raised ALT or AST before dosing narrows the margin for interpreting any post-dose rise and was an exclusion in the first-in-human trial, which admitted only volunteers free of clinically significant disease.

  • Sex: No sex difference in adverse events has been reported for this class. The relevant asymmetry is reproductive: the first-in-human trial excluded women able to become pregnant, so exposure risk in pregnancy is entirely uncharacterized.

  • Pre-existing cardiac and renal disease: The phase 2 protocol excludes New York Heart Association class III or IV heart failure, ejection fraction under 30%, uncontrolled arrhythmia, and blood pressure above 160/100 mmHg. Safety in decompensated disease is unknown.

  • Age: Human exposure spans ages 18–55 in phase 1 and 18–80 in phase 2. Adults beyond 80, in whom polypharmacy and reduced kidney clearance are common, have no data at all.

Key Interactions & Contraindications

  • Prescription lipid drugs (atorvastatin, rosuvastatin, ezetimibe, bempedoic acid): No pharmacokinetic interaction is expected, since gene-silencing RNA is not a CYP substrate. Severity: none. Consequence: additive lipid lowering only, with statin therapy continued rather than substituted.

  • PCSK9 inhibitors (drugs that help the liver clear more cholesterol from the blood, such as evolocumab, alirocumab, inclisiran): These themselves lower Lp(a) by roughly a fifth. Severity: monitor. Consequence: additive suppression and a harder-to-interpret Lp(a) result, addressed by staggering measurement rather than the doses.

  • Anticoagulants and antiplatelets (warfarin, apixaban, clopidogrel, aspirin): No metabolic interaction. Severity: caution, theoretical. Consequence: because apolipoprotein(a) impedes clot breakdown, removing it may add to bleeding tendency; no bleeding excess has been observed in class trials.

  • Other CYP-metabolized medicines (amiodarone, ketoconazole, ritonavir, grapefruit juice as an inhibitor source): Severity: none identified. Consequence: no dose adjustment; sugar-conjugated gene-silencing drugs neither inhibit nor induce these enzymes and have no meaningful transporter handling.

  • Supplements that also lower Lp(a) (niacin, L-Carnitine, N-acetylcysteine, omega-3 fatty acids, coenzyme Q10 as a neutral comparator): Severity: monitor. Consequence: additive lowering plus niacin’s flushing and glucose effects, mitigated by an eight-week separation between any new supplement and a Kylo-11 measurement window.

  • Other interventions (lipoprotein apheresis, hormone therapy): Severity: caution. Consequence: apheresis and estrogen both lower Lp(a) independently, confounding attribution of effect, which sequencing agreed with the trial investigator resolves better than unmonitored combination.

Populations who should avoid Kylo-11:

  • Anyone outside a registered clinical trial — the compound is not approved for supply in any jurisdiction
  • Pregnant or breastfeeding women, and women able to become pregnant not using protocol-grade contraception
  • Adults under 18 or, on current data, over 80
  • Heart failure of New York Heart Association class III or IV, or left ventricular ejection fraction below 30%
  • Uncontrolled hypertension at or above 160 mmHg systolic or 100 mmHg diastolic
  • Any malignancy within the preceding 5 years, other than treated non-melanoma skin cancer, cervical carcinoma in situ, breast ductal carcinoma in situ, or stage 1 prostate carcinoma
  • Active hepatitis B, hepatitis C, human immunodeficiency virus, or syphilis infection
  • Severe hepatic impairment (Child-Pugh class C), given hepatocyte-directed delivery

Risk Mitigation Strategies

  • Trial-only access: Obtaining Kylo-11 through any channel other than an enrolled site removes dose verification, adverse-event reporting, and follow-up, mitigating nothing and adding counterfeit exposure to a compound with no antidote.

  • Lowest studied dose first: The dose-ranging trial spanned 9 mg to 600 mg. Choosing the lowest dose that reaches the Lp(a) target limits the twelve-month unremovable exposure that a higher dose commits a person to.

  • Liver panel before dosing and at 4 and 12 weeks: Detects the transaminase elevation expected from hepatocyte-directed delivery early enough to distinguish it from unrelated causes before the next annual injection is considered.

  • Apolipoprotein B alongside every lipid panel: Prevents the misreading risk, in which an Lp(a)-driven fall in measured LDL cholesterol prompts an unwarranted reduction of statin or ezetimibe therapy.

  • Injection-site rotation and cold compression: Reduces the injection-site redness, pain, and swelling characteristic of subcutaneous sugar-conjugated gene-silencing drugs, which are dose-local rather than systemic.

  • A documented pre-treatment Lp(a) on a single assay platform: Guards against the mitigation failure of judging response against an incomparable baseline, since mass-based and particle-based assays disagree.

Therapeutic Protocol

  • No established practitioner protocol: Kylo-11 is investigational. The only protocols in existence are the two sponsor trial protocols; no clinic prescribes it, and no dosing consensus exists outside them.

  • Trial dosing schedule: Phase 1 gave one subcutaneous injection at 9, 30, 75, 225, 450, or 600 mg. Phase 2 tests three undisclosed dose levels against placebo in 204 participants with ASCVD.

  • Competing approach — maximal apolipoprotein B lowering: The conventional route accepts Lp(a) as unmodifiable and drives every other artery-damaging particle down with statins, ezetimibe, and PCSK9 inhibitors, an approach associated with Cleveland Clinic and Steven Nissen’s trial program.

  • Competing approach — lipoprotein apheresis: Filtering the particle out of the blood every one to two weeks, developed in German lipid centers and advanced by Patrick Moriarty at the University of Kansas, remains the only approved route.

  • Competing approach — antisense blockade: Sotirios Tsimikas at the University of California, San Diego, developed the antisense route that reached outcome testing first. None of these three is the default; each trades convenience against evidence differently.

  • Time of day: Not a consideration. The injection is given once and works through a target whose production has no clinically relevant daily rhythm; trial protocols specify no dosing time.

  • Half-life: Plasma half-life is hours, since unbound drug is cleared by nucleases and the kidney. The effect half-life is months, set by intracellular persistence — which is why annual dosing is even proposed.

  • Single versus split dosing: Single dose, always. Splitting would defeat the design; the whole premise is that one loaded dose sustains silencing for a year, and no split-dose arm has been studied.

  • Genetic influence on dose: LPA kringle IV type-2 repeat number determines baseline Lp(a) and therefore the dose needed to reach a target. No pharmacogenetic dosing rule exists, and CYP-linked variants are irrelevant to this class.

  • Sex differences in dosing: None established. Trials enrolled both sexes without dose stratification; the only sex-specific rule is the contraception requirement, not a dose adjustment.

  • Age considerations: Phase 1 capped enrollment at 55 years and phase 2 at 80. Nothing supports dosing beyond 80, where reduced kidney clearance of unbound drug is plausible but unstudied.

  • Baseline biomarker gate: Entry required Lp(a) above 75 nmol/L, with a dedicated cohort above 200 nmol/L. Below that threshold there is neither a tested dose nor a rationale.

  • Pre-existing conditions: Advanced heart failure, uncontrolled arrhythmia, recent malignancy, active viral hepatitis, and uncontrolled hypertension all exclude a person from the studied population and therefore from any protocol claim.

Discontinuation & Cycling

  • Intended as lifelong: The target is a genetically fixed particle that returns to baseline once silencing lapses, so the design intent is indefinite annual re-dosing rather than a defined course.

  • Discontinuation is passive: Stopping means declining the next injection. There is no taper, no antidote, and no way to hasten recovery; Lp(a) simply climbs back as liver cells clear the loaded complex.

  • No withdrawal syndrome: None has been reported for this class. The only rebound is the return of Lp(a) toward its inherited set point, which is a return to baseline risk rather than an overshoot.

  • Recovery is slow: Off-treatment follow-up of a same-class agent shows Lp(a) still substantially suppressed a year after the final dose (O’Donoghue et al.), so a missed annual dose is not an immediate loss of effect.

  • Cycling is not applicable: No tolerance, receptor downregulation, or loss of potency on repeat dosing has been described for gene-silencing RNA, so deliberate drug holidays would forfeit effect without gaining anything.

Sourcing and Quality

  • No legitimate consumer supply exists: Kylo-11 has no marketing authorization anywhere. It cannot be bought from a pharmacy, imported legally for personal use, or compounded, and it is not available off-label.

  • Trial enrollment is the only route: Supply comes from the sponsor to registered sites under a clinical protocol. Sites for the phase 2 study span China and the United States, including California, Florida, and other states.

  • Gray-market “research grade” RNA strands are a different product: Vendors selling siRNA sequences supply unformulated, unconjugated material of unverified identity, without the sugar ligand that creates liver selectivity, and with no sterility assurance.

  • Third-party testing is not meaningful here: Independent certificates of analysis exist for supplements, not for investigational biologics. Identity, potency, and endotoxin control for Kylo-11 rest entirely on the sponsor’s manufacturing quality system.

  • Formulation considerations: The product is a sterile aqueous solution for subcutaneous injection, supplied and stored by the trial site. No consumer-facing formulation, strength, or excipient information has been published.

Practical Considerations

  • Time to effect: Lp(a) begins falling within two to four weeks of a single injection and reaches its lowest point around weeks 8 to 16, based on the class pattern and the phase 2 endpoint window of weeks 8 to 26.

  • Regulatory status: Investigational everywhere. No approval, no off-label pathway, and no expanded access program has been announced; the phase 1 record notes the trial was not conducted under an FDA (Food and Drug Administration) regulated-drug designation.

  • Common pitfall — treating an isolated Lp(a) number: Deep Lp(a) lowering has never been shown to prevent an event. Treating the number as if the outcome were established is the central error available in this field today.

  • Common pitfall — mismatched assay units: Results reported in mg/dL and nmol/L are not interconvertible by a fixed factor, because mass-based assays are distorted by particle size. Comparing across platforms manufactures false response or false failure.

  • Cost and access: Trial participation is free; commercial pricing is unknown. Approved injectable lipid biologics run into thousands of dollars per year against pennies per day for generic statins.

  • Structural financial bias: That cost gap gives insurers and national health systems a systematic incentive to favor generics, which shapes reimbursement, guideline thresholds, and which comparisons get funded — a bias running opposite to the manufacturers’ own.

Interaction with Foundational Habits

  • Sleep: No interaction, direct or indirect. The injection has no stimulant, sedative, or hormonal action, no daily dosing to time around bedtime, and no sleep-related adverse events have been reported for gene-silencing RNA drugs targeting the liver. Nothing about a sleep protocol needs to change around dosing.

  • Nutrition: Essentially none in either direction, which is the striking part. Lp(a) is largely diet-independent; notably, replacing saturated fat with unsaturated fat lowers LDL cholesterol yet tends to raise Lp(a) slightly, and very-low-carbohydrate patterns can raise it too. Neither meaningfully alters what a gene-silencing injection achieves.

  • Exercise: No interaction with the drug, and almost none with the target — endurance and resistance training barely move Lp(a). Exercise remains additive for overall arterial risk through blood pressure, insulin sensitivity, and fitness, so it addresses different risk channels rather than potentiating or blunting this one.

  • Stress management: No direct interaction. Lp(a) behaves as a mild acute-phase reactant, rising transiently during inflammatory illness or major surgery, so measuring a level during such a period can mislead. Lp(a) testing is therefore deferred until at least four weeks after any acute illness.

Monitoring Protocol & Defining Success

A genuine baseline before any dose consists of two Lp(a) measurements on the same assay platform, in particle units, taken at least four weeks apart and at least four weeks clear of acute illness, since a single reading during inflammation misleads. These are paired with apolipoprotein B, a full lipid panel, liver enzymes with bilirubin, kidney function, glycated hemoglobin, platelet count, and — for women of reproductive age — a pregnancy test. All of these mirror the entry and exclusion criteria of the two Kylo-11 trials rather than any approved label.

Thereafter, the cadence follows the drug’s kinetics rather than a conventional schedule: liver enzymes at 4 and 12 weeks, Lp(a) at 12 weeks to capture the nadir, then Lp(a) with apolipoprotein B and liver enzymes at 6 months and 12 months, and annually before each subsequent injection.

Biomarker Optimal Functional Range Why Measure It? Context/Notes
Lipoprotein(a) Below 75 nmol/L; below 30 nmol/L preferred The direct target; defines both eligibility and response A particle-number assay in nmol/L on one platform throughout is the usable form; mg/dL results are size-distorted and not interconvertible. Conventional labs often flag only above 125 nmol/L
Apolipoprotein B Below 80 mg/dL; below 60 mg/dL with established arterial disease Counts every artery-damaging particle, so it cannot be flattered by Lp(a) removal Fasting not required. Conventional labs flag only above 90–130 mg/dL, far above the functional target. Best paired with the lipid panel to detect the falsely reassuring LDL fall
LDL cholesterol Below 70 mg/dL Tracks the accounting shift caused by removing Lp(a)-carried cholesterol Conventional reference ranges permit up to 100 mg/dL, well above the functional target here. Interpretable only alongside apolipoprotein B
ALT and AST Below 25 U/L (men below 30 U/L) Detects the transaminase rise expected from hepatocyte-directed delivery ALT and AST are alanine and aspartate aminotransferase, enzymes released when liver cells are stressed. Conventional labs tolerate up to 40 U/L. Drawn together with bilirubin
Total bilirubin 0.3–1.0 mg/dL Separates a benign enzyme rise from genuine hepatic impairment Gilbert syndrome, an inherited benign cause of raised bilirubin, was a trial exclusion, and identifying it before dosing avoids misattribution
eGFR Above 90 mL/min/1.73 m² Unbound drug is partly cleared by the kidney eGFR is estimated glomerular filtration rate, a calculated measure of kidney filtering. Conventional labs call 60 normal; that is a low bar here
hs-CRP Below 1.0 mg/L Flags the inflammation that transiently inflates an Lp(a) reading hs-CRP is high-sensitivity C-reactive protein, a general marker of body-wide inflammation. Conventional labs call anything below 3.0 mg/L normal. Lp(a) testing is deferred while it is elevated
HbA1c 5.0–5.4% Watches the speculative metabolic signal attached to very low Lp(a) HbA1c is glycated hemoglobin, reflecting average blood glucose over three months. Conventional labs accept up to 5.6%
Platelet count 175–250 ×10⁹/L Screens for the platelet decline seen with some nucleic-acid drug classes Not reported as a problem for gene-silencing RNA, but a low-cost precaution given the annual dosing interval and absence of an antidote. The conventional reference range is far wider at 150–450 ×10⁹/L

Qualitative markers worth tracking alongside the laboratory work:

  • Injection-site comfort — redness, pain, or swelling in the days after dosing, and whether it recurs at the next injection
  • Energy and exertional tolerance — any change in how far or how hard exercise can be pushed without chest discomfort or breathlessness
  • Sleep quality and general wellbeing, as a non-specific early-warning indicator for anything the laboratory panel does not cover
  • Confidence in the plan — whether the twelve-month commitment to an unwithdrawable effect still feels proportionate at each annual decision point

Emerging Research

  • First-in-human dose-ranging study: NCT06363851 enrolled 71 healthy volunteers with Lp(a) above 75 nmol/L across six single doses from 9 to 600 mg, with adverse-event incidence to week 24 as its primary endpoint. Completed December 2025 and published in August 2026.

  • Phase 2 dose-finding study in established disease: NCT07327840 is randomizing 204 participants with ASCVD and elevated Lp(a) across three dose levels against placebo, with time-averaged percent change in Lp(a) over weeks 8–26 as primary endpoint. Primary completion February 2027.

  • The outcome question, first reading: NCT04023552 tested an antisense agent in 8,323 participants for major cardiovascular events. It completed in July 2026 with no public readout at the time of writing; a null result would weaken the case for every agent in this space, including Kylo-11.

  • The outcome question, siRNA reading: NCT05581303 is running olpasiran in 7,297 participants with ASCVD, with primary completion in March 2028 — the first event-driven test of the same silencing mechanism Kylo-11 uses.

  • The largest bet on the mechanism: NCT06292013 is testing lepodisiran in 17,300 adults through March 2029, in both secondary and primary prevention. Its size and breadth make it the study most likely to settle whether Lp(a) lowering prevents anything.

  • Duration as the competitive axis: Off-treatment work such as The Off-Treatment Effects of Olpasiran on Lipoprotein(a) Lowering: OCEAN(a)-DOSE Extension Period Results — O’Donoghue et al., 2024 — defines how long silencing persists. If class duration proves comparable, Kylo-11’s once-yearly premise loses its differentiator; if not, it gains one.

  • Safety questions that could weaken the case: No study has asked whether decades of near-total Lp(a) suppression carries a cost. Lipoprotein(a) concentrations, isoform size, and risk of type 2 diabetes: a Mendelian randomisation study — Kamstrup & Nordestgaard, 2013 — frames the hypothesis a long-term extension would test.

  • Measurement standardization: Assay disagreement between mass and particle units, discussed in Considerations for routinely testing for high lipoprotein(a) — Nurmohamed et al., 2023, a narrative review — still limits cross-trial comparison and could distort how Kylo-11 reads against competitors.

  • Who funds the next answers: Recommendations that Lp(a) be measured once in a lifetime originate with professional cardiology societies whose members conduct, and whose meetings and research are funded by, the industry programs developing these drugs, including Kylo-11’s.

Conclusion

Kylo-11 is an experimental injection that silences the liver gene behind lipoprotein(a), a blood particle whose level is fixed largely at birth and which standard cholesterol treatment barely touches. Its distinguishing claim is duration: one injection, an effect measured in months rather than weeks, and a stated ambition of once-a-year dosing.

What is established is narrow. Deep and durable lowering of the particle has now been observed and published, though in a small first-in-human study of healthy volunteers funded by the company that owns the compound and stands to profit from its approval. Everything else in this review — about hardened arteries, about the heart valve, about fewer events — is reasoning from inherited genetics and from other drugs aimed at the same target, none of which has yet shown that lowering this particle prevents anything.

The safety picture is correspondingly thin. Local injection reactions are the expected nuisance of this drug family, though none occurred in the one study of this compound, and the same long duration that makes the drug attractive means an unwanted effect could not be withdrawn for about a year. Guidance encouraging people to measure this particle at all comes from cardiology societies whose members and research funding are tied to the companies developing these drugs, while insurers face the opposite pull toward cheap generics.

For someone who already knows their level is high and has exhausted the conventional levers, the interest here is real, and entirely forward-looking.

Top - Benefits - Risks - Protocol