Utility-based dose finding: BOIN12 and its extensions

A utility of the response and the toxicity

The BOIN12 design of Lin, Zhou, Yan, Li and Yuan (2020) looks for the optimal biological dose (OBD) by scoring the four outcomes a patient can have. A response without a dose-limiting toxicity (DLT) scores 100 and a DLT without a response 0; a patient with neither scores psi00 and a patient with both psi11, 40 and 60 by default. The utility of a dose is the expected score of its patients. After each cohort, the boundaries of the BOIN design rule out an escalation from a dose that looks too toxic, and the next cohort goes to the most desirable dose among the current dose and its neighbors, the desirability being the posterior probability that the utility of the dose exceeds a benchmark. At the end of the trial the OBD is the admissible dose, up to the estimated MTD, with the highest estimated utility.

The package provides four designs of this family:

The trials and the selection of the OBD run in C++. In these functions target is the highest acceptable DLT rate and target_resp the lowest acceptable response rate. The target_resp of the BOIN-ET functions, vignette("simFastBOIN-boin-et"), is a different quantity, the response rate that BOIN-ET aims at.

The desirability score table

boin12_decision_table() tabulates, for every number of patients, DLTs and responses at a dose, whether the dose is admissible and the rank of its desirability among the outcomes of the table, the score of Table 3 of Lin et al. (2020).

tab <- boin12_decision_table(target = 0.35, target_resp = 0.25)
tab_3 <- as.data.frame(tab)[tab$n == 3, ]
tab_3[, c("n", "n_tox", "n_resp", "admissible", "score")]
#>    n n_tox n_resp admissible score
#> 2  3     0      0       TRUE    35
#> 3  3     0      1       TRUE    55
#> 4  3     0      2       TRUE    76
#> 5  3     0      3       TRUE    91
#> 6  3     1      0       TRUE    24
#> 7  3     1      1       TRUE    44
#> 8  3     1      2       TRUE    63
#> 9  3     1      3       TRUE    80
#> 10 3     2      0       TRUE    13
#> 11 3     2      1       TRUE    31
#> 12 3     2      2       TRUE    48
#> 13 3     2      3       TRUE    69
#> 14 3     3      0      FALSE    NA
#> 15 3     3      1      FALSE    NA
#> 16 3     3      2      FALSE    NA
#> 17 3     3      3      FALSE    NA

With the default numbers of patients, 0, 3, 6 and 9, the table has 166 rows, the cells of Table 3 of Lin et al. (2020); the lines above are those for three patients. A higher score means a more desirable dose, and only the order of the scores of the doses compared matters for the next cohort. With three DLTs among three patients the dose is inadmissible whatever the responses: under a uniform prior, the posterior probability that its DLT rate exceeds 0.35 is then 0.985, above the cutoff 0.95.

Operating characteristics for one scenario

Scenario 1 of Lin et al. (2020) has five doses, an upper DLT limit of 0.35, a lower response limit of 0.25 and twelve cohorts of three patients. oc_scenarios() gives the true utility of every dose and the true MTD and OBD.

p_true <- c(0.05, 0.12, 0.27, 0.35, 0.50)
p_resp <- c(0.20, 0.35, 0.36, 0.37, 0.38)

truth_1 <- oc_scenarios(list("Scenario 1" = list(p_true = p_true,
                                                 p_resp = p_resp)),
                        target = 0.35, target_resp = 0.25)
truth_1
#> True scenarios: 1 scenario(s), 5 doses 
#> True MTD: DLT probability closest to 0.35 
#> True OBD: highest utility among the doses with DLT probability <= 0.35 and response probability >= 0.25 
#> Utilities: 100 (response, no DLT), 40 (neither), 60 (both), 0 (DLT, no response)
#> 
#> Scenario 1
#>             1    2    3    4    5
#> p_true   0.05 0.12 0.27 0.35 0.50
#> p_resp   0.20 0.35 0.36 0.37 0.38
#> utility  50.0 56.2 50.8 48.2 42.8
#> true MTD                   *     
#> true OBD         *

The response rate hardly increases beyond dose 2 while the DLT rate does, so the OBD, dose 2, is well below the MTD, dose 4.

oc <- sim_boin12(target = 0.35, target_resp = 0.25, p_true = p_true,
                 p_resp = p_resp, n_cohort = 12, cohort_size = 3,
                 n_trials = 2000, seed = 123)
oc
#> BOIN12 operating characteristics
#>   DLT limit            : 0.35
#>   response limit       : 0.25
#>   trials               : 2000
#>   cohorts              : 12 of size 3
#>   max sample size      : 36
#>   lambda_e / lambda_d  : 0.2763 / 0.4189
#>   utility scores       : 100, 40, 60, 0; benchmark 70.5
#>   ties / no candidate  : higher / nearest
#>   exploration from     : any
#>   OBD selection        : utility by posterior_mean
#> 
#>                         DL1  DL2  DL3  DL4  DL5 Total / No OBD
#> True DLT rate (%)       5.0 12.0 27.0 35.0 50.0               
#> True response rate (%) 20.0 35.0 36.0 37.0 38.0               
#> OBD selected (%)       21.4 46.6 21.4  9.2  1.3            0.0
#> Patients treated        7.7 12.0  8.5  5.6  2.3           36.0
#> Patients with DLT       0.4  1.5  2.2  1.9  1.2            7.2
#> Patients with response  1.6  4.1  3.1  2.1  0.9           11.7
#> 
#> Trials stopped early: 0%

Dose 2 is selected most often, in 46.6 percent of the trials, and 0 percent of the trials select no OBD.

The rules of the simulations of the article

The defaults follow the BOIN12 software. Three arguments select the rules under which Lin et al. (2020) simulated the design, where they differ: tie_break = "random" breaks ties of desirability at random instead of choosing the higher dose, no_candidate = "stop" ends the trial without an OBD when no dose allowed for the next cohort is admissible instead of going to the nearest admissible dose, and explore_from = "admissible" applies the rule that escalates after 9 patients when the next dose is untried only at an admissible dose. The article states the first; the other two are not written in it and were inferred from the published results, which they reproduce (see the last section).

oc_article <- sim_boin12(target = 0.35, target_resp = 0.25, p_true = p_true,
                         p_resp = p_resp, n_cohort = 12, cohort_size = 3,
                         n_trials = 2000, tie_break = "random",
                         no_candidate = "stop", explore_from = "admissible",
                         seed = 123)
selection <- rbind(default = c(oc$sel_percent, oc$percent_no_obd),
                   article = c(oc_article$sel_percent,
                               oc_article$percent_no_obd))
colnames(selection) <- c(paste0("DL", 1:5), "None")
round(selection, 1)
#>          DL1  DL2  DL3 DL4 DL5 None
#> default 21.4 46.6 21.4 9.2 1.3  0.0
#> article 19.4 47.5 21.3 9.6 1.8  0.3

TITE-BOIN12

When the DLT and the response take long to assess, TITE-BOIN12 (Zhou et al., 2022) treats new patients while earlier ones are still followed. A pending outcome is replaced by its expectation given the time followed, as in Equation (11) of the article (pending_estimate = "conditional", the default); pending_estimate = "marginal" replaces it by the estimated rate itself, which reproduces the desirabilities of Table S.3 of their Supplement. While fewer than half of the patients at the current dose have finished both assessments, the next decision waits (min_completed_share = 0.5). The simulations of the article have windows of 45 days for the DLT and 60 days for the response and three patients a month; the unit of time is the day.

tite <- sim_tite_boin12(target = 0.35, target_resp = 0.25, p_true = p_true,
                        p_resp = p_resp, n_cohort = 12, cohort_size = 3,
                        window = 45, resp_window = 60, accrual_rate = 3 / 30,
                        n_trials = 2000, seed = 123)
tite
#> TITE-BOIN12 operating characteristics
#>   DLT limit            : 0.35
#>   response limit       : 0.25
#>   trials               : 2000
#>   cohorts              : 12 of size 3
#>   max sample size      : 36
#>   lambda_e / lambda_d  : 0.2763 / 0.4189
#>   utility scores       : 100, 40, 60, 0; benchmark 70.5
#>   ties / no candidate  : higher / nearest
#>   exploration from     : any
#>   OBD selection        : utility by posterior_mean
#> 
#>                         DL1  DL2  DL3  DL4  DL5 Total / No OBD
#> True DLT rate (%)       5.0 12.0 27.0 35.0 50.0               
#> True response rate (%) 20.0 35.0 36.0 37.0 38.0               
#> OBD selected (%)       18.1 47.2 21.3 11.1  2.1            0.1
#> Patients treated        7.4 12.1  8.4  5.5  2.5           36.0
#> Patients with DLT       0.4  1.4  2.2  1.9  1.2            7.2
#> Patients with response  1.5  4.3  3.0  2.0  0.9           11.7
#> 
#> Trials stopped early: 0.1%
#> Average trial duration: 563

When every decision waits for all the outcomes and the patients arrive at fixed intervals, the trials are those of BOIN12 with the same seed, and only the time differs.

wait <- sim_tite_boin12(target = 0.35, target_resp = 0.25, p_true = p_true,
                        p_resp = p_resp, n_cohort = 12, cohort_size = 3,
                        window = 45, resp_window = 60, accrual_rate = 3 / 30,
                        accrual = "fixed", min_completed_share = 1,
                        n_trials = 2000, seed = 123)
all.equal(wait$sel_percent, oc$sel_percent)
#> [1] TRUE
round(c(TITE = tite$duration, wait = wait$duration) / 30, 1)
#> TITE wait 
#> 18.8 30.6

The trials last 18.8 months on average with TITE-BOIN12 and 30.6 months when every decision waits.

Comb-BOIN12 and TITE-Comb-BOIN12

Comb-BOIN12 (Lu et al., 2025) applies the utility, the desirability and the admissibility of BOIN12 to a matrix of combinations of two drugs, with the levels of drug A in the rows and those of drug B in the columns. The next cohort goes to a combination chosen by the desirability among the current one and its neighbors along both drugs, as BOIN12 does along the doses. At the end the response rates are estimated by a quadratic logistic model, and the admissible combination with the highest utility up to the MTD combination is selected. The utility of the article, pE - wT * pT, is that of BOIN12 with psi00 = 100 * wT / (1 + wT) and psi11 = 100 / (1 + wT), and the weight w_u of its benchmark is benchmark_weight. Scenario 16 of the article has two levels of drug A and three of drug B.

p_true_ab <- matrix(c(0.05, 0.10, 0.15, 0.15, 0.20, 0.45), nrow = 2)
p_resp_ab <- matrix(c(0.10, 0.20, 0.30, 0.40, 0.50, 0.60), nrow = 2)

comb_design <- list(target = 0.33, target_resp = 0.2, p_true = p_true_ab,
                    p_resp = p_resp_ab, n_cohort = 14, cohort_size = 3,
                    psi00 = 100 * 0.11 / 1.11, psi11 = 100 / 1.11,
                    benchmark_weight = 0.6, extrasafe = TRUE, offset = 0.10,
                    n_trials = 1000, seed = 123)
oc_comb <- do.call(sim_comb_boin12, comb_design)
oc_comb
#> Comb-BOIN12 operating characteristics
#>   DLT limit            : 0.33
#>   response limit       : 0.2
#>   dose levels          : 2 of drug A, 3 of drug B
#>   trials               : 1000
#>   cohorts              : 14 of size 3
#>   max sample size      : 42
#>   lambda_e / lambda_d  : 0.2604 / 0.3947
#>   utility scores       : 100, 9.9099, 90.0901, 0; benchmark 54.79
#>   ties / no candidate  : higher / nearest
#>   OBD selection        : logistic response, rectangle bound
#> 
#>                    Item Drug A   B1   B2   B3
#>       True DLT rate (%)     A1  5.0 15.0 20.0
#>                             A2 10.0 15.0 45.0
#>  True response rate (%)     A1 10.0 30.0 50.0
#>                             A2 20.0 40.0 60.0
#>        OBD selected (%)     A1  5.3 13.1 36.6
#>                             A2  3.0 25.2 15.8
#>        Patients treated     A1  4.0  5.7 11.3
#>                             A2  3.4  9.1  8.2
#>       Patients with DLT     A1  0.2  0.9  2.3
#>                             A2  0.3  1.4  3.7
#>  Patients with response     A1  0.4  1.7  5.6
#>                             A2  0.7  3.7  4.9
#> 
#> No OBD selected                      : 1%
#> Trials stopped early                 : 1%
#> Average number of patients           : 41.6
#> 
#> True OBD (highest true utility)      : A1B3
#>   Selected as the OBD                : 36.6%
#>   Patients treated there             : 27.2%
#> Combinations above the DLT limit     : 1 of 6
#>   Selected as the OBD                : 15.8%
#>   Patients treated there             : 19.6%

The true OBD combination is A1B3, and it is selected in 36.6 percent of the trials.

The article states that the OBD combination is chosen among the combinations up to the levels of both drugs of the MTD combination (obd_bound = "rectangle", the default). The published selection percentages agree better with candidates bounded, at every level of drug B, by the MTD along drug A (obd_bound = "contour"), a rule inferred from the published results. The same trials give

oc_contour <- do.call(sim_comb_boin12,
                      c(comb_design, list(obd_bound = "contour")))
labels <- paste0("A", rep(1:2, 3), "B", rep(1:3, each = 2))
bounds <- rbind(rectangle = as.vector(oc_comb$sel_percent),
                contour = as.vector(oc_contour$sel_percent))
colnames(bounds) <- labels
round(bounds, 1)
#>           A1B1 A2B1 A1B2 A2B2 A1B3 A2B3
#> rectangle  5.3  3.0 13.1 25.2 36.6 15.8
#> contour    1.0  2.5  4.0 23.0 48.3 20.2

TITE-Comb-BOIN12 applies the rules of Comb-BOIN12 at the decision times of TITE-BOIN12. By default the pending outcomes count in the DLT rate of the decision, in the admissibility and in the utility, as in TITE-BOIN12; Sections 2.2 and 3.1 of Lu et al. (2025) state that they count in the utility only, the decision and the admissibility using the observed outcomes (pending_use = "utility"). The article does not describe the accrual or the waiting rule, for which the package uses those of TITE-BOIN12. One of the settings of Figure 2 of the article has windows of one month for the DLT and two months for the response and two patients a month, here with uniform times to the events, as Section 2.2 of the article assumes:

tite_comb_design <- c(comb_design[names(comb_design) != "n_trials"],
                      list(window = 1, resp_window = 2, accrual_rate = 2,
                           event_time = "uniform", n_trials = 500))
oc_tc <- do.call(sim_tite_comb_boin12, tite_comb_design)
oc_tc_utility <- do.call(sim_tite_comb_boin12,
                         c(tite_comb_design, list(pending_use = "utility")))
pending <- rbind(all = c(as.vector(oc_tc$sel_percent), oc_tc$duration),
                 utility = c(as.vector(oc_tc_utility$sel_percent),
                             oc_tc_utility$duration))
colnames(pending) <- c(labels, "Months")
round(pending, 1)
#>         A1B1 A2B1 A1B2 A2B2 A1B3 A2B3 Months
#> all      3.2  2.6 12.0 26.2 38.0 17.8   26.2
#> utility  2.8  3.0 16.2 27.2 37.2 12.8   26.0

Comparing designs

oc_scenarios(), oc_results() and oc_compare() set the results of different designs side by side, also for designs that look for the MTD. Scenarios 1 to 3 of Lin et al. (2020) are run with BOIN, which selects the MTD, and with BOIN-ET and BOIN12, which select an OBD. BOIN-ET needs the response rate it aims at, set here to 0.40, and its default correlation of the DLT and the response, 0.2, is set to 0, the default of BOIN12, so that the designs are compared on the same outcomes.

scenarios <- list(
  "Scenario 1" = list(p_true = c(0.05, 0.12, 0.27, 0.35, 0.50),
                      p_resp = c(0.20, 0.35, 0.36, 0.37, 0.38)),
  "Scenario 2" = list(p_true = c(0.05, 0.12, 0.27, 0.35, 0.50),
                      p_resp = c(0.01, 0.05, 0.30, 0.60, 0.60)),
  "Scenario 3" = list(p_true = c(0.03, 0.05, 0.20, 0.22, 0.45),
                      p_resp = c(0.05, 0.10, 0.50, 0.68, 0.70))
)
truth <- oc_scenarios(scenarios, target = 0.35, target_resp = 0.25)

boin <- sim_boin_multi(target = 0.35, scenarios = scenarios, n_cohort = 12,
                       cohort_size = 3, n_trials = 2000, seed = 123)
boin_et <- sim_boin_et_multi(target = 0.35, target_resp = 0.40,
                             scenarios = scenarios, n_cohort = 12,
                             cohort_size = 3, n_trials = 2000, resp_cor = 0,
                             seed = 123)
boin12 <- sim_boin12_multi(target = 0.35, target_resp = 0.25,
                           scenarios = scenarios, n_cohort = 12,
                           cohort_size = 3, n_trials = 2000, seed = 123)

comparison <- oc_compare(BOIN = boin, "BOIN-ET" = boin_et, BOIN12 = boin12,
                         truth = truth)
comparison
#> Comparison of 3 design(s) with the true doses
#> Correct: the true MTD for designs selecting an MTD, the true OBD for designs selecting an OBD
#> Overdose: true DLT probability above 0.35 
#> 
#> Scenario 1
#>               Correct (%) Pts at correct (%) Overdose sel. (%) Pts at overdose
#> BOIN (MTD)           42.6               25.2              12.1             3.3
#> BOIN-ET (OBD)        45.4               47.4               1.4             0.5
#> BOIN12 (OBD)         46.6               33.2               1.3             2.3
#>               No selection (%) Patients
#> BOIN (MTD)                 0.0     33.8
#> BOIN-ET (OBD)              0.0     36.0
#> BOIN12 (OBD)               0.0     36.0
#> 
#> Scenario 2
#>               Correct (%) Pts at correct (%) Overdose sel. (%) Pts at overdose
#> BOIN (MTD)           42.6               25.2              12.1             3.3
#> BOIN-ET (OBD)        42.5               27.7               0.9             0.3
#> BOIN12 (OBD)         47.1               29.9               4.2             3.0
#>               No selection (%) Patients
#> BOIN (MTD)                 0.0     33.8
#> BOIN-ET (OBD)              0.2     36.0
#> BOIN12 (OBD)               4.2     35.7
#> 
#> Scenario 3
#>               Correct (%) Pts at correct (%) Overdose sel. (%) Pts at overdose
#> BOIN (MTD)           40.0               24.8              40.0             8.6
#> BOIN-ET (OBD)        18.0               10.8               0.3             0.1
#> BOIN12 (OBD)         60.7               35.6               8.5             3.8
#>               No selection (%) Patients
#> BOIN (MTD)                 0.0     34.5
#> BOIN-ET (OBD)              0.0     36.0
#> BOIN12 (OBD)               0.8     36.0

The correct selection counts the true MTD for BOIN and the true OBD for the other two designs; an overdose is a dose with a DLT rate above 0.35. In Scenario 1, where the OBD lies below the MTD, BOIN12 selects the OBD in 46.6 percent of the trials and BOIN-ET in 45.4 percent. The figure shows the same table.

plot(comparison)

How the implementation is checked

The scripts in inst/reproduce/ and inst/validation/ compare the package with the articles and with the ‘escalation’ package, and write tables of judgments in their results/ folders, which are summarized here. A simulated value passes when it lies within three or four standard errors of the difference between the two simulations, as each script states, plus the rounding of the published value. Values listed for information are not judged.

References

Lin, R., Zhou, Y., Yan, F., Li, D. and Yuan, Y. (2020). BOIN12: Bayesian optimal interval phase I/II trial design for utility-based dose finding in immunotherapy and targeted therapies. JCO Precision Oncology, 4, 1393-1402.

Zhou, Y., Lin, R., Lee, J. J., Li, D., Wang, L., Li, R. and Yuan, Y. (2022). TITE-BOIN12: A Bayesian phase I/II trial design to find the optimal biological dose with late-onset toxicity and efficacy. Statistics in Medicine, 41(11), 1918-1931.

Lu, M., Zhang, J., Yuan, Y. and Lin, R. (2025). Comb-BOIN12: A utility-based Bayesian optimal interval design for dose optimization in cancer drug-combination trials. Statistics in Biopharmaceutical Research, 17(2), 266-276.