An S4 generic method providing a more abstract interface to trial simulation than
the simulate_trials function of package escalation.
This abstraction is needed to support simulations in which escalation's simple
vectors of ‘true toxicity probabilities’ are replaced by precautionary's
more realistic toxicity-distribution generators.
simulate_trials(
selector_factory,
num_sims,
true_prob_tox,
true_prob_eff = NULL,
...
)
# S4 method for selector_factory,numeric,hyper_mtdi_distribution,ANY
simulate_trials(
selector_factory,
num_sims,
true_prob_tox,
true_prob_eff = NULL,
...
)
# S4 method for selector_factory,numeric,mtdi_distribution,ANY
simulate_trials(
selector_factory,
num_sims,
true_prob_tox,
true_prob_eff = NULL,
...
)
# S4 method for exact,missing,mtdi_distribution,ANY
simulate_trials(
selector_factory,
num_sims,
true_prob_tox,
true_prob_eff = NULL,
...
)
# S4 method for exact,numeric,mtdi_distribution,missing
simulate_trials(
selector_factory,
num_sims,
true_prob_tox,
true_prob_eff = NULL,
...
)
# S4 method for exact,numeric,hyper_mtdi_distribution,missing
simulate_trials(
selector_factory,
num_sims,
true_prob_tox,
true_prob_eff = NULL,
...
)An object of S3 class selector_factory
Number of simulations to run
A generator of toxicity distributions
Provided for compatibility with simulate_trials
Passed to subroutines
If invoked interactively with num_sims > 10, then a
txtProgressBar is displayed in the console. The condition on
num_sims has the useful side effect of allowing this function
to be invoked iteratively by extend (with num_sims = 10)
without the nuisance of nested progress bars.
old <- options(dose_levels = c(0.5, 1, 2, 4, 6, 8))
mtdi_gen <- hyper_mtdi_lognormal(CV = 1
, median_mtd = 6, median_sdlog = 0.5
, units="mg/kg")
num_sims <- ifelse(interactive()
, 300
, 15 # avoid taxing CRAN servers
)
hsims <- get_three_plus_three(num_doses = 6,
allow_deescalate = TRUE) %>%
simulate_trials(
num_sims = num_sims
, true_prob_tox = mtdi_gen)
summary(hsims, ordinalizer=NULL) # vanilla summary with binary toxicity
#> # A tibble: 7 × 7
#> dose `dose (mg/kg)` tox n true_prob_tox prob_recommend prob_administer
#> <ord> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 NoDose 0 0 0 0 0 0
#> 2 1 0.5 0 3 0 0 0.158
#> 3 2 1 0.133 3.4 0 0.133 0.179
#> 4 3 2 0.6 4.2 0 0.267 0.221
#> 5 4 4 0.933 4.2 0 0.333 0.221
#> 6 5 6 1.2 3 0 0 0.158
#> 7 6 8 0.133 1.2 0 0.267 0.0632
summary(hsims, ordinalizer = function(dose, r0 = sqrt(2))
c(Gr1=dose/r0^2, Gr2=dose/r0, Gr3=dose, Gr4=dose*r0, Gr5=dose*r0^2)
)
#> $escalation
#> # A tibble: 7 × 7
#> dose `dose (mg/kg)` tox n true_prob_tox prob_recommend prob_administer
#> <ord> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 NoDose 0 0 0 0 0 0
#> 2 1 0.5 0 3 0 0 0.158
#> 3 2 1 0.133 3.4 0 0.133 0.179
#> 4 3 2 0.6 4.2 0 0.267 0.221
#> 5 4 4 0.933 4.2 0 0.333 0.221
#> 6 5 6 1.2 3 0 0 0.158
#> 7 6 8 0.133 1.2 0 0.267 0.0632
#>
#> $safety
#> None Gr1 Gr2 Gr3 Gr4 Gr5 Total
#> 13 1 1 2 1 1 19
#>
#> $toxTab
#> Tox
#> rep None Gr1 Gr2 Gr3 Gr4 Gr5 Total
#> 1 16 2 0 0 0 0 18
#> 2 11 0 3 2 2 0 18
#> 3 19 1 2 2 0 0 24
#> 4 9 3 0 2 1 0 15
#> 5 10 1 0 1 3 0 15
#> 6 18 0 0 0 0 0 18
#> 7 10 4 2 3 1 1 21
#> 8 10 1 4 1 1 1 18
#> 9 12 1 1 1 1 2 18
#> 10 9 2 0 3 0 1 15
#> 11 18 0 1 2 0 0 21
#> 12 16 1 0 1 1 2 21
#> 13 18 2 2 1 1 0 24
#> 14 14 1 0 2 1 0 18
#> 15 12 1 3 4 0 1 21
#>
hsims <- hsims %>% extend(num_sims = num_sims)
summary(hsims, ordinalizer = function(dose, r0 = sqrt(2))
c(Gr1=dose/r0^2, Gr2=dose/r0, Gr3=dose, Gr4=dose*r0, Gr5=dose*r0^2)
)$safety
#> None Gr1 Gr2 Gr3 Gr4 Gr5 Total
#> 12 1 1 2 1 1 18
# Set a CRM skeleton from the average probs in above simulation
num_sims <- ifelse(interactive()
, 16
, 4 # avoid taxing CRAN servers
)
get_dfcrm(skeleton = hsims$avg_prob_tox
,target = 0.25
) %>% stop_at_n(n = 24) %>%
simulate_trials(
num_sims = num_sims
, true_prob_tox = mtdi_gen
) -> crm_hsims
summary(crm_hsims
, ordinalizer = function(MTDi, r0 = sqrt(2))
MTDi * r0^c(Gr1=-2, Gr2=-1, Gr3=0, Gr4=1, Gr5=2)
)
#> $escalation
#> # A tibble: 7 × 7
#> dose `dose (mg/kg)` tox n true_prob_tox prob_recommend prob_administer
#> <ord> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 NoDose 0 0 0 0 0 0
#> 2 1 0.5 0.25 3.75 0 0 0.156
#> 3 2 1 0 0.75 0 0 0.0312
#> 4 3 2 1.25 6 0 0.5 0.25
#> 5 4 4 2.5 12 0 0.25 0.5
#> 6 5 6 0 0 0 0.25 0
#> 7 6 8 1.25 1.5 0 0 0.0625
#>
#> $safety
#> None Gr1 Gr2 Gr3 Gr4 Gr5 Total
#> 15 2 2 2 2 1 24.0
#>
#> $toxTab
#> Tox
#> rep None Gr1 Gr2 Gr3 Gr4 Gr5 Total
#> 1 14 2 3 3 2 0 24
#> 2 12 1 3 3 1 4 24
#> 3 18 1 1 1 3 0 24
#> 4 16 2 2 3 0 1 24
#>
options(old)
old <- options(dose_levels = c(2, 6, 20, 60, 180, 400))
mtdi_dist <- mtdi_lognormal(CV = 0.5
,median = 140
,units = "ng/kg/week")
num_sims <- ifelse(interactive()
, 100
, 10 # avoid taxing CRAN servers
)
sims <- get_three_plus_three(num_doses = 6) %>%
simulate_trials(
num_sims = num_sims
, true_prob_tox = mtdi_dist)
# Now set a proper ordinalizer via options():
options(ordinalizer = function(dose, r0) {
c(Gr1=dose/r0^2, Gr2=dose/r0, Gr3=dose, Gr4=dose*r0, Gr5=dose*r0^2)
})
summary(sims, r0=2)
#> $escalation
#> # A tibble: 7 × 7
#> dose `dose (ng/kg/week)` tox n true_prob_tox prob_recommend
#> <ord> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 NoDose 0 0 0 0 0
#> 2 1 2 0 3 1.19e-19 0
#> 3 2 6 0 3 1.30e-11 0
#> 4 3 20 0 3 1.90e- 5 0
#> 5 4 60 0.2 3.6 3.64e- 2 0.9
#> 6 5 180 2.2 3.3 7.03e- 1 0.1
#> 7 6 400 0.3 0.3 9.87e- 1 0
#> # … with 1 more variable: prob_administer <dbl>
#>
#> $safety
#> None Gr1 Gr2 Gr3 Gr4 Gr5 Total
#> 9 2 2 2 1 0 16
#>
#> $toxTab
#> Tox
#> rep None Gr1 Gr2 Gr3 Gr4 Gr5 Total
#> 1 9 1 3 1 1 0 15
#> 2 9 3 1 0 2 0 15
#> 3 9 3 2 4 0 0 18
#> 4 9 2 2 1 1 0 15
#> 5 10 2 3 0 1 2 18
#> 6 9 3 1 1 1 0 15
#> 7 8 1 3 1 2 0 15
#> 8 10 4 3 3 1 0 21
#> 9 8 2 2 3 0 0 15
#> 10 9 1 3 2 0 0 15
#>
# Set a CRM skeleton from the average probs in above simulation
get_dfcrm(skeleton = sims$true_prob_tox
,target = 0.25
) %>% stop_at_n(n = 24) %>%
simulate_trials(
num_sims = 20
, true_prob_tox = mtdi_dist
) -> crm_sims
summary(crm_sims
, ordinalizer = function(MTDi, r0 = sqrt(2))
MTDi * r0^c(Gr1=-2, Gr2=-1, Gr3=0, Gr4=1, Gr5=2)
)
#> $escalation
#> # A tibble: 7 × 7
#> dose `dose (ng/kg/week)` tox n true_prob_tox prob_recommend
#> <ord> <dbl> <dbl> <dbl> <dbl> <dbl>
#> 1 NoDose 0 0 0 0 0
#> 2 1 2 0 3 1.19e-19 0
#> 3 2 6 0 0 1.30e-11 0
#> 4 3 20 0 0 1.90e- 5 0
#> 5 4 60 0.6 17.0 3.64e- 2 0.95
#> 6 5 180 2.55 4.05 7.03e- 1 0.05
#> 7 6 400 0 0 9.87e- 1 0
#> # … with 1 more variable: prob_administer <dbl>
#>
#> $safety
#> None Gr1 Gr2 Gr3 Gr4 Gr5 Total
#> 14 4 3 2 1 1 24.0
#>
#> $toxTab
#> Tox
#> rep None Gr1 Gr2 Gr3 Gr4 Gr5 Total
#> 1 13 8 2 0 1 0 24
#> 2 13 2 4 3 1 1 24
#> 3 14 6 1 2 1 0 24
#> 4 15 4 3 1 0 1 24
#> 5 16 3 2 2 1 0 24
#> 6 14 3 3 2 1 1 24
#> 7 14 3 4 1 1 1 24
#> 8 16 3 1 2 1 1 24
#> 9 15 3 3 2 0 1 24
#> 10 9 6 4 2 1 2 24
#> 11 14 5 3 0 1 1 24
#> 12 12 4 6 1 1 0 24
#> 13 16 5 1 2 0 0 24
#> 14 14 5 3 2 0 0 24
#> 15 16 4 1 2 1 0 24
#> 16 17 5 1 1 0 0 24
#> 17 9 4 5 2 2 2 24
#> 18 10 2 6 2 1 3 24
#> 19 17 4 1 0 2 0 24
#> 20 13 3 4 1 3 0 24
#>
if (interactive()) { # don't overtax CRAN servers
crm_sims <- crm_sims %>% extend(target_mcse = 0.1)
summary(crm_sims
, ordinalizer = function(MTDi, r0 = sqrt(2))
MTDi * r0^c(Gr1=-2, Gr2=-1, Gr3=0, Gr4=1, Gr5=2)
)$safety
}
options(old)
old <- options(
dose_levels = c(0.5, 1, 2, 4, 6),
ordinalizer = function(MTDi, r0 = 1.5) {
MTDi * r0 ^ c(Gr1=-2, Gr2=-1, Gr3=0, Gr4=1, Gr5=2)
})
mtdi_dist <- mtdi_lognormal(CV = 2
,median = 5
,units = "mg/kg")
design <- get_three_plus_three(num_doses = 5, allow_deescalate = TRUE)
# Note use of wrapper function 'exact'; see ?precautionary::exact.
exact(design) %>% simulate_trials(true_prob_tox = mtdi_dist) -> EXACT
stopifnot(all.equal(1, sum(exp(EXACT$log_pi))))
summary(EXACT)$safety
#>
#> None Gr1 Gr2 Gr3 Gr4 Gr5 Total
#> [1,] 10.04956 1.591247 1.364949 1.088902 0.8066095 1.317687 16.21896
if (interactive()) { # don't overtax CRAN servers
# Compare with result of discrete-event simulation
design %>% simulate_trials(
num_sims = 200
, true_prob_tox = mtdi_dist
) -> SIMS
summary(SIMS)$safety
}
options(old)
old <- options(dose_levels = c(0.5, 1, 2, 4, 6, 8))
mtdi_gen <- hyper_mtdi_lognormal(CV = 1
, median_mtd = 6, median_sdlog = 0.5
, units="mg/kg")
options(ordinalizer = function(dose, r0 = sqrt(2))
c(Gr1=dose/r0^2, Gr2=dose/r0, Gr3=dose, Gr4=dose*r0, Gr5=dose*r0^2)
)
design <- get_three_plus_three(num_doses = 6, allow_deescalate = TRUE)
ehsims <- simulate_trials(
exact(design)
, num_sims = 50
, true_prob_tox = mtdi_gen
)
summary(ehsims)$safety
#> None Gr1 Gr2 Gr3 Gr4
#> Expected participants 11.8616115 2.086239 1.7305842 1.20890367 0.73808846
#> MCSE 0.3243027 0.115837 0.1030522 0.05547655 0.02814529
#> Gr5 Total
#> Expected participants 0.93582514 18.5612521
#> MCSE 0.07391789 0.3542685
options(old)