At single-pulse 1,000-2,000 mg dose (the Mayo regimen):

• Day 1-2 (dosing days): Most users report nothing acutely. A subset report mild GI upset (nausea, loose stools) — likely a high-polyphenol-load effect, not specific. Occasional mild headache. Some users report a vague sensation of "calm" or "easier sleep" within hours of dose; this is small and not consistently replicated across users.
• Day 3-7 (post-pulse window): A subset of users report a "settling" or "anti-inflammatory wash" — joints feel less stiff, gum or skin inflammation eases, fatigue lifts mildly. This is the most-reported subjective signal, but it's also where placebo runs strongest and where the cellular-cleanup story would not predict same-week subjective effects.
• Long-term (months): Bryan Johnson-style daily dosers report no specific subjective benefit attributable to fisetin within their stack. Athletes report no detectable change in training quality or recovery time at typical 100-500 mg/day dosing.

At daily 100-500 mg dose:

• Subjective effects are essentially nil for most users — consistent with the pharmacokinetic problem.
• The "Bryan Johnson stack" rationale is biomarker-driven, not feel-driven.
• High-dose daily use (1+ g/day) is uncommon and not supported by senolytic mechanism (cells need to be pulsed, not pressured).

Why so little subjective signal?

1. Bioavailability bottleneck. Plasma free fisetin after 100 mg oral is nanomolar — well below most cellular-binding affinities. The drug is doing little until you get to gram-scale pulses with fat co-ingestion.
2. Cellular mechanism, not neurochemical. Senescent-cell clearance, NF-κB downregulation, and SIRT1 activation are slow, downstream, biomarker-level changes. They don't manifest as "I feel sharper today" the way modafinil or selank does.
3. Population doesn't have the substrate. A 20-year-old has virtually no senescent-cell burden. There's nothing for the senolytic to clear. The same dose in an 80-year-old with high p16+ load would (in theory) produce more biomarker movement.

Honest variability: ~70% of users feel nothing identifiable. ~20% report a mild post-pulse "wash" subjectively similar to a hard sauna session or a fasting-mimicking diet completion. ~5-10% report mild GI/headache adverse effects. Almost no one reports a "noticeable cognitive boost" — and reports that do exist are likely placebo or food-related (the strawberry connection makes Bayesian-priors generous).

• Tolerance: None documented. Fisetin's mechanisms (senolytic, SIRT1, NF-κB) don't have classical receptor-desensitization pathways like opioid or stimulant receptors. The pulse protocol itself is built around long inter-pulse intervals (28+ days), which prevent any pharmacodynamic adaptation.
• Senescent-cell substrate depletion: Theoretically, sequential pulses clear the senescent-cell pool, and subsequent pulses have less substrate. This is the mechanism by which "hit-and-run" makes sense — once you've cleared the cells, the next pulse can be deferred. The pool replenishes slowly (months to years in mice), so monthly pulses likely exceed substrate availability after a few cycles; bimonthly or quarterly may be more appropriate after initial 3-6 month load.
• Cycling pattern (Mayo): 2 days on, 28 days off, repeat ~3-12 cycles. No data on long-term cycling > 1 year.
• Reset: Not needed in any conventional sense — fisetin doesn't down-regulate anything that needs to recover.

Synergistic with

• Quercetin — most-combined-with fisetin (n=67 in community data). Both flavonols, both senolytic-leaning. Dasatinib + quercetin (D+Q) is the other major senolytic protocol (Hickson 2019); fisetin + quercetin is a "two-flavonol senolytic" attempt without dasatinib's hematologic risk. Mechanistically plausible; no head-to-head trial.
• Rapamycin — both modulate mTOR. Common in longevity-focused stacks. Theoretical concern: both suppress mTOR signaling → additive immune/metabolic effects. Communities run this combo without obvious problems; clinical trials avoid it (probably for endpoint-disentanglement reasons, not safety).
• NAD+ precursors (NMN, NR) — orthogonal mechanism. Bryan-Johnson-style stacks combine routinely.
• Resveratrol — overlapping SIRT1 activation; redundant on that mechanism but additive on antioxidant. Mostly stacked for branding-of-longevity rather than synergy.
• Spermidine — autophagy-stimulating; complementary to senolytic. Found together in many longevity protocols.
• NAC — antioxidant overlap; helpful for glutathione support during high-pulse polyphenol load.
• Omega-3 (EPA/DHA) — anti-inflammatory overlap; co-administering with fisetin pulse may improve absorption (lipid carrier) and broaden inflammation knockdown.

Avoid stacking with (or extra caution)

• Warfarin or other anticoagulants — CYP2C9 inhibition by fisetin theoretically raises warfarin levels. Monitor INR more frequently around pulses.
• Antiplatelet agents (clopidogrel, aspirin) — theoretical additive bleeding risk; fisetin's antiplatelet signal is mild but real.
• Statins (especially rosuvastatin) — OATP inhibition by fisetin reduces statin uptake; statin efficacy could fall. Timing-separation by 4-6 h during pulse days.
• Tamoxifen / estrogen-receptor-modulator therapy — fisetin's CYP1B1 inhibition + estrogenic-pathway modulation creates an uncharacterized interaction. Defer to oncology.
• Active chemotherapy — senescent cells play roles in some chemo responses (treatment-induced senescence is part of mechanism). Senolytic interference during chemo is contraindicated absent oncology specialist sign-off.
• MAO-B inhibitors (selegiline, rasagiline) — fisetin's weak in-vitro MAO inhibition signal raises theoretical additive concern. Low-confidence; relevant mostly at very high doses.

Neutral / safe co-administration

• The user's V4 stack (Mg, NAC, citicoline, PS, DHA, curcumin, rhodiola, theanine, glycine, D3/K2, beta-alanine, vitamin C) — all neutral or mildly synergistic.
• Creatine, beta-alanine, electrolytes — neutral.
• Modafinil, selank, semax — no published interactions; mechanistically distinct.
• BPC-157, TB-500 — peptide tissue-repair stack, no documented interaction. Note theoretical concern about senolytic-during-repair timing (see Side effects).

Metabolic profile:

• CYP2C9 inhibition (moderate, in vitro). Theoretical impact on warfarin, phenytoin, ibuprofen, glipizide.
• CYP3A4 inhibition (mild, in vitro). Theoretical impact on alprazolam, clonazepam, midazolam, statins (atorvastatin, simvastatin), apixaban, sildenafil, immunosuppressants.
• CYP2C19 inhibition (mild). Theoretical impact on clopidogrel activation (REDUCED efficacy — clopidogrel needs 2C19 to activate; this is the opposite-direction concern from other CYP interactions).
• CYP1B1 inhibition. Relevant to estradiol metabolism — may shift estrogen metabolite balance.
• OATP1A2 / OATP2B1 inhibition. Reduces uptake of OATP-substrate drugs (rosuvastatin, fexofenadine, certain ACEi/ARB).
• P-glycoprotein inhibition. Theoretical impact on digoxin, dabigatran, apixaban.
• 5α-reductase inhibition (mild). Theoretical interaction with testosterone-to-DHT conversion, finasteride, dutasteride.

The honest picture: Almost all of the published "fisetin-drug interaction" literature is in vitro or molecular docking, not human clinical. The community-data AI-seeded interaction block on this page flags ~20 theoretical interactions, almost all rated "low" or "speculative" confidence. The clinically actionable subset:

1. Warfarin — check INR around pulses (moderate concern, MODERATE confidence).
2. Aspirin / antiplatelet — additive bleeding theoretical (low-moderate concern).
3. Statins (rosuvastatin specifically) — OATP-mediated uptake reduction (moderate confidence in mechanism).
4. Active CYP3A4 substrates with narrow therapeutic index (immunosuppressants, certain antiarrhythmics) — defer to prescriber.

For Dylan specifically: zero current Rx. No interaction concern. If modafinil is added (V5 plan) — modafinil's primary metabolic route is amide hydrolysis, not CYP, so fisetin co-administration is mechanistically clean. If post-injury opioid (CYP3A4 substrate) is ever needed, defer fisetin pulse to post-recovery.

• CYP2C9 *2 / *3 alleles (poor metabolizers, ~10-20% of Caucasians). Theoretically more sensitive to fisetin's CYP2C9 inhibition + already lean on 2C9 for warfarin / NSAIDs / sulfonylureas. Relevant if poly-pharmacy emerges later.
• CYP3A5 expresser status (rs776746). Modulates 3A-substrate clearance; fisetin's 3A inhibition more impactful in expressers (who otherwise clear 3A substrates faster).
• GSTM1 / GSTT1 deletions. Affect phase-II detox of phenolic compounds. GSTM1-null users may experience slower fisetin clearance + altered conjugate profile. No clinical relevance demonstrated.
• NRF2 (NFE2L2) polymorphisms. Variants in the Nrf2 promoter could modulate fisetin's antioxidant-pathway induction. Not characterized clinically.
• TP53 / CDKN2A variants (senescence biology). Variants affecting p53/p16 baseline senescence load could theoretically modulate senolytic response. Speculative.

For Dylan (23andMe pending June 2026): No fisetin-specific PGx call to make pre-data. Post-results, primary check is CYP2C9 phenotype if any future warfarin / chronic NSAID / sulfonylurea exposure is on the table. Senolytic response PGx is not yet a clinically actionable field.

Sourcing-difficulty: Easy. OTC, no regulatory friction. No WADA listing. The choice is between cheap-generic-for-daily vs Novusol-for-pulse — same molecule, different absorption.

Cost for the recommended protocol (Dylan, 1-2 pulses/year): 1-2 × (2 days × 1,500 mg × Novusol form) ≈ $40-80/year. Trivial relative to V4 stack budget.

Baseline (before starting any fisetin protocol)

• hs-CRP — primary inflammation readout; sub-2 mg/L target.
• IL-6 — more sensitive than CRP; <2 pg/mL target.
• TNF-α — broader SASP marker; <8 pg/mL.
• CBC + diff — particularly neutrophil/lymphocyte ratio (NLR) as a coarse inflammaging marker.
• CMP — liver enzymes (ALT/AST) baseline before polyphenol load; kidney function.
• Lipid panel — for the population where statin co-administration is on the table.
• INR — if on warfarin.
• p16INK4a (research only) — peripheral blood T-cell p16 has been used in trial populations as a senescence biomarker. Not commercially available; epigenetic-clock services (TruDiagnostic, etc.) overlap.

During use (pulse protocol)

• Subjective tracking only on pulse days — GI tolerance, sleep, mood.
• Re-check hs-CRP + IL-6 4 weeks after pulse #1, then quarterly. Looking for trend reductions over 6+ months; single-pulse changes are noise.
• Annual: Full CBC + CMP + lipid + hs-CRP + IL-6.

Optional advanced (for serious longevity quantification)

• GlycanAge / TruDiagnostic Horvath clock — annual; epigenetic age estimate.
• Bristle / oral inflammation — if doing dental/gum tracking.
• Grip strength — proxy for muscle senescence (Murray 2025 supports muscle-specific fisetin effect).