Table 1.

Time to different stages of competitive release of T− cells for varying simulated treatment regimens

Time to progression	5% T−	90% T−	% Dose
Representative patient #1—with “cheater” population
MTD	3950	4621	100%
Metronomic	4556	4950	24.1%
Adaptive	8557	Indefinite	1.0%
Representative patient #2—without “cheater” population
MTD	323	784	100%
Metronomic	323	784	24.1%
Adaptive	323	Indefinite	< 1.0%

No T− population is present before treatment in Patient #1 and only arises after therapy is given. Both MTD and the long induction of metronomic therapy results in a T− population appearing and it rapidly comprises the whole tumor. There is a slight lengthening of control in Patient #1 with metronomic therapy. Total dosing declines 24% relative to MTD. Under adaptive therapy, the substantial population of T+ cells delays the establishment of T− cells and provides durable control for a long period with minimal amounts of drug. Adaptive therapy prevents full competitive release of T−. Patient #2 has a population of T− in the tumor prior to first treatment. This is shown at time 323 in all cases. Both MTD and the long induction of metronomic therapy allows the T− population to quickly rise and experience full competitive release. Adaptive therapy in patient #2 prevents full competitive release of the T− population. The significantly lower percentage of drug given in the modeled adaptive therapies compared to the clinical trial percentages is due to the on/off pharmacokenetics in the model that leads to immediate reponses of cell populations. Patients likely experience more gradual changes and can only see a shift in therapy at most every 4 weeks