Fundamental Treatment Levers

Treating cancer as an attractor fact forces four levers: re-flip the basin (differentiation), restore the barrier, remove the drive, restore surveillance. All four are now measured as stochastic trajectories, and cytotoxic relapse is an explicit regrowth curve — the malignant fraction regrows when the basin is intact, decays when it is emptied. APL and CAR-T validate two levers. Grade [V]/[L].

All four levers are measured as stochastic trajectories. Differentiation (Lever A) empties the malignant basin as the drive crosses the spinodal (residual ON occupancy 0.90→0.00, non-cytotoxic); drive removal (Lever C) leaves a committed population occupied (1.00, relapse) but a healthy one prevented (0.00). Barrier restoration (Lever B) collapses the counted crossing rate ~7× as the barrier is restored; surveillance restoration (Lever D) clears the reservoir as a time course; and cytotoxic relapse is an explicit regrowth curve recovering to 100% of capacity while differentiation decays to ≈0.

Malignancy is a basin, not a headcount

The deep claim of this volume is that a tumour is a region of the attractor landscape, not merely a number of cells. A durable cure must change the landscape — empty or destabilise the malignant basin — rather than only reducing the count of cells sitting in it.

Four levers, all from one kernel

Lever A — basin re-flip (differentiation). A malignant cell re-flips into a healthy basin once a differentiating drive reaches the spinodal, with no cytotoxicity. The clinical anchor is acute promyelocytic leukaemia, the first acute leukaemia cured by a non-cytotoxic, differentiating regimen (all-trans retinoic acid plus arsenic trioxide).

Lever B — barrier restoration. Restoring tumour-suppressor or epigenetic control raises the barrier, and the Kramers rate collapses exponentially — a large multiplicative drop in crossing rate for a modest barrier gain.

Lever C — drive removal (etiologic). Removing the carcinogen drive prevents un-committed crossings, but by hysteresis it does not reverse an already-committed cell. It is preventive, not curative once committed — matching smoking-cessation epidemiology, where future risk falls while an established tumour persists.

Lever D — surveillance restoration. This volume's own seam. Lowering the immune-escape factor clears the committed reservoir at a fixed crossing-rate and, because the factor is a common multiplier, acts across every site. The anchor is CAR-T and checkpoint blockade, curative in refractory leukaemia and lymphoma.

Why cytotoxic-only relapses

Cytotoxic killing removes cells but leaves the barrier and the basin unchanged, so any surviving malignant cell stays malignant and the basin refills — hysteretic relapse. The kernel therefore predicts that the cleanest durable cures are attractor- and field-level and non-cytotoxic; the two cleanest real haematologic cures (differentiation in APL, surveillance in CAR-T) are exactly Lever A and Lever D.

Levers A and C emerge as measured trajectories

Levers A and C are no longer read off a single deterministic settle; they are measured from direct stochastic simulations of the substrate. For Lever A a population starts in the malignant basin and a differentiating drive is applied: the residual malignant-basin occupancy collapses from 0.90 to 0.00 as the drive crosses the spinodal (measured 0.5-crossing at 0.81× the spinodal, approached from below because thermal activation helps the cell over the shrinking barrier), with no cytotoxicity. The re-flip threshold is therefore the spinodal, measured organ-by-organ.

For Lever C the carcinogen drive is removed under noise. A committed population stays in the malignant basin (occupancy 1.00 — hysteresis, the substrate origin of relapse) while a healthy population stays out of it (occupancy 0.00 — prevention), so removing the cause is preventive but not curative. Cytotoxic killing is the same measurement: it leaves the landscape intact, so its survivors keep occupancy 1.00 and the basin refills. Only differentiation drives occupancy to 0.00 and empties the basin — the measured reversal-vs-relapse contrast, with the absolute agent dose and schedule left [O].

Levers B and D emerge as measured trajectories

Lever B — barrier restoration. The crossing-rate collapse is no longer read off the closed-form exponential; it is measured. As a fraction of the carcinogen-eroded barrier is stepped back up, the counted malignant crossing rate falls monotonically — a 7.1× multiplicative drop at full restoration — and the logarithm of the measured rate is linear in the restored barrier (R²=0.994, Arrhenius slope -7.4 recovering −1/D=-8.3), so the Kramers collapse emerges in the therapeutic direction rather than being assumed.

Lever D — surveillance restoration. The reservoir clearance is measured as a time course. Starting from a high-escape reservoir, restoring surveillance makes the counted committed reservoir decay toward a floor that falls with deeper surveillance; floor × surveillance is constant, so the floor scales as 1/(1−escape) — the steady T10 seam recovered as a trajectory endpoint — and, rescaled by influx, the AML and lymphoma floors collapse onto the same 1/surveillance curve (cross-site gap 0.025). Deeper surveillance also clears the reservoir faster.

Cytotoxic relapse is a measured regrowth curve

The relapse failure mode is made explicit as a population trajectory gated by the measured R19 basin fraction. After cytotoxic killing the survivors remain in the malignant basin (measured ON fraction 1.00) and the malignant fraction regrows toward the carrying capacity, recovering to 100% — relapse. After a differentiating re-flip the basin is emptied (measured ON fraction 0.00) and the malignant fraction decays to 0% — a cure, with no cytotoxicity. Sweeping the kill fraction shows the decisive point: a deeper kill only lengthens the regrowth delay (time-to-half 0.0 → 4.7 as the kill rises 50%→99%) while every depth still recovers fully. Relapse is determined by the basin, not by how many cells are killed — maximal cytotoxic kill does not cure, the attractor does.

Combination therapy converts relapse into cure

The relapse curve and the cure curve can be put on the same axes and combined. A cytotoxic cull alone drives the malignant fraction down at first but, because the basin is untouched, it regrows to the carrying capacity (final 100% of K — relapse). A single basin-acting lever already cures (differentiation alone and restored surveillance both decay to ≈0). The decisive measurement is the combination: adding the cytotoxic cull to either basin lever keeps the cure (final ≈0) and lowers the cumulative burden — the area under N(t)/K falls from 0.95 (differentiation alone) to 0.09 with a cull added, and from 0.64 (surveillance alone) to 0.09 — so the cull accelerates the approach without changing the destination.

The control is explicit and honest: the same cull alone never cures at any depth — only a basin/niche lever converts relapse to cure. And the surveillance channel shows an emergent threshold at the growth rate: a sub-threshold surveillance (μ=0.3) relapses to an interior fixed point at the predicted N*=K(1−μ/r) (measured 70% of K), while a supra-threshold surveillance (μ=1.0, 2.0) clears to a cure (6.4%, 0.0% of K). The conversion, the cull's burden reduction, the cull-alone insufficiency and the growth-rate threshold are measured; the absolute rates and the clinical schedule stay [O].

What the kernel does not claim

VP does not invent these therapies. It re-derives why they are the fundamental class and names the cytotoxic relapse failure mode mechanically. The kernel predicts a class of intervention and the relapse mode; it does not predict a molecule, a dose, a schedule, or an individual patient's response, which require pharmacology and biomarkers outside the deterministic substrate — stated [O].