1 Introduction
This document describes the FACTS XML file (“.facts”) file format for FACTS Staged Multiple Endpoint.
It is intended for expert FACTS users who wish to create modified versions of “.facts” files using scripts. A full and deep understanding of FACTS Core and Staged designs is assumed.
1.1 IDs
There are number of parameters that IDs, such as “guid”, “SetId”, “qoi_id” … these are Microsoft “unique” identifiers, there should be no need to change these. It does not matter if the same Ids are used in successive “.facts” files, for instance if creating a range of sub-versions of a single original file.
1.2 Defaults
FACTS stores a super-set of the FACTS parameters in the “.facts” file. There are so many parameters in the FACTS file that we cannot check the behavior of FACTS if some of these are missing, so are advice is to include *all* the parameters even if not required. The only parameters that it is clear can be omitted are those in the optional parameter sets: BaselineParameterSet and LongitudinalParameterSet that are not required if the respective features are not enabled.
1.3 Parameter Order
FACTS processes the parameters in the “.facts” file in the order in which the appear in the file. Some parameters depend on other parameters to have been already defined, so it is recommended that the order given here (the order FACTS itself uses when writing out the parameters) is retained. For many parameters changing their order should not be a problem, but we are not going to define, or guarantee, which those are.
2 The values in a “.facts” file
2.1 Numbers of items
ACS: The number of studies used for the hierarchical prior for the active comparator
ADC: The dose in a 2D dose model design which have been included in the study
CF: The number of column factors in a 2D dose model design
CS: The number of studies used for the hierarchical prior for the control
D: The number of doses (or treatment arms) in the design
DC: The number of possible dose combinations in a 2D dose model design
DQ: The number of Decision QOIs
I: The number of interims (when defined by information
LM: The number of longitudinal models
PostQ: The number of Posterior Probability QOIs
PredQ: The number of Predictive Probability QOIs
PvQ: The number of p-value QOIs
R: The number of regions in an accrual profile
RF: The number of row factors in a 2D dose model design
TQ: The number of target QOIs
V: The number of patient visits
Var: The number of variants
LM*V: This is the value used when a different prior is used at different visits. The values are stored as lists of lists (nested <item>s). The outer dimension is the number of LM and the inner dimension the number of visits. If the priors are common across all models, just the first lists of values for visits is used, the others are present but ignored.
2.2 Common structures
2.2.1 Gamma prior
The parameters are stored in a <gamma> value as <mu> – the central value of the prior, and <n> - the weight of the prior in equivalent number of observations. E.g.:
<gamma>
<mu>10</mu>
<n>1</n>
</gamma>2.2.2 Normal prior
The parameters are stored in a <menasdpair> value as <mean> - the mean of the prior and <sd> - the standard deviation. E.g.:
<meansdpair>
<mean>0</mean>
<sd>10</sd>
</meansdpair>2.2.3 Beta prior
The parameters are stored in a <betadist> value as <alpha> - the alpha parameter of the beta distribution prior and <beta> - the beta parameter of the beta distribution prior. E.g.:
<betadist>
<alpha>2</alpha>
<beta>2</beta>
</betadist>2.2.4 Posterior probability QOI
The parameters stored in a <post_prob_qoi> structure are:
condition: 3=greater than, 4=less than
tte_qoi_type: leave at 1 for non-tte endpoints
relation
is_absolute: “True” or “False”
comparison arm: id is_absolute=False, this is the name of the comparison arm, the name must appear as a name in the <dose_names> property.
delta: the value of the delta in the comparison
qoi_parametrization_type: 0=standard parametrization, 1=log-odds parametrization (dichotomous endpoint only), 2=hazard rate parametrization (TTE endpoint only).
alt_name: the short name for the QOI, the default QOIs have specific names, the user added QOIs have names of the form QOI_1_N for the nth posterior probability QOI
is_default_qoi: “True” or “False”, if the QOI is a default QOI its values should not be changed.
qoi_id: this is a Microsoft unique id, so that the QOI can be referenced from other parameters where the QOI is used.
Example:
<post_prob_qoi>
<condition>3<\condition>
<tte_qoi_type>1<tte_qoi_type>
<relation>
<is_absolute>False</is_absolute>
<comparison_arm>Control</comparison_arm>
<delta>1</delta>
</relation>
<qoi_parametrization_type>0</qoi_parametrization_type>
<alt_name>Pr(CSD)</alt_name>
<is_default_qoi>True</is_default_qoi>
<qoi_id>ac13283c-dffc-47d6-9ba1-a371a633e626</qoi_id>
<\post_prob_qoi>2.2.5 Predictive probability QOI
The parameters stored in a <pred_prob_qoi> structure are:
pred_prob_type: 1=Bayesian, 2=Conditional Power
phase: 1=Current trial, 2=Future trial
test_type: 1=superiority, 2=non-inferioirty (Future Trial only)
subjects_per_arm: number of subjects arm (Future Trial only)
margin: Superiority/Non-inferiority margin (Future Trial only)
dropout_type: 0=LOCF, 1=BOCF, 2=Ignore, 3=Ignore (Current Trial only)
freq_test_type: 1=Unadjusted, 2=Bonferroni, 3=Dunnett (Current Trial only)
sample_size: 1=Current Enrollment, 2=Trial Maximum Enrollment (Current Trial only) (note: Trial Maximm cannot be used with adaptive allocation)
alpha: one-sided alpha
alt_name: the short name for the QOI, the default QOIs have specific names, the user added QOIs have names of the form QOI_2_N for the nth predictive probability QOI
is_default_qoi: “True” or “False”, if the QOI is a default QOI its values should not be changed.
qoi_id: this is a Microsoft unique id, so that the QOI can be referenced from other parameters where the QOI is used.
Example:
<pred_prob_qoi>
<pred_prob_type>1</pred_prob_type>
<phase>2</phase>
<test_type>1</test_type>
<subjects_per_arm>250</subjects_per_arm>
<margin>0</margin>
<alpha>0.025</alpha>
<alt_name>Pr(SS)</alt_name>
<is_default_qoi>True</is_default_qoi>
<qoi_id>2556a9c6-7277-4437-865c-5e5b7d451634</qoi_id>
</pred_prob_qoi>
<pred_prob_qoi>
<pred_prob_type>1</pred_prob_type>
<phase>1</phase>
<dropouts_type>0</dropouts_type>
<freq_test_type>1</freq_test_type>
<sample_size>2</sample_size>
<alpha>0.027</alpha>
<alt_name>QOI_2_3</alt_name>
<is_default_qoi>False</is_default_qoi>
<qoi_id>39c533a1-0cfd-476f-9561-aea191730f91</qoi_id>
</pred_prob_qoi>2.2.6 P-value QOI
The parameters stored in a <pvalue_ qoi> structure are:
pvalue_dropout_type: 0=LOCF, 1=BOCF, 2=Ignore, 3=Ignore (Current Trial only)
test_type: 1=Unadjusted, 2=Bonferroni, 3=Dunnett, 4=Trend Test
alt_name: the short name for the QOI, the default QOIs have specific names, the user added QOIs have names of the form QOI_3_N for the nth pvalue QOI
is_default_qoi: “True” or “False”, if the QOI is a default QOI its values should not be changed.
qoi_id: this is a Microsoft unique id, so that the QOI can be referenced from other parameters where the QOI is used.
Example:
<pvalue_qoi>
<pvalue_dropouts_type>0</pvalue_dropouts_type>
<test_type>2</test_type>
<alt_name>QOI_3_2</alt_name>
<is_default_qoi>False</is_default_qoi>
<qoi_id>44ba7e29-da1f-4de8-913b-24aa0ed6b3c9</qoi_id>
</pvalue_qoi>2.2.7 Target Dose QOI
The parameters stored in a <target_dose_qoi> structure are:
target_type: 1=Max, 2=MED, 3=EDq
tte_qoi_type: leave at 1 for non-tte endpoints
dose_combo_range_type: when using the 2D dose model, this represents the chosen 2D dose grid range type for which to evaluate the target dose QOIs. 0=Full Range, 1=Only columns, 2=Only rows.
dose_combo_range_arm: when using the 2D dose model, this represents the specific column or row (dependent on whether dose_combo_range_type is equal to 1 or 2, respectively) of the 2D dose grid for which to evaluate the target dose QOIs.
relation (MED & EDq target types only, present but empty for Max target type)
is_absolute: “True” or “False”
comparison arm: id is_absolute=False, this is the name of the comparison arm, the name must appear as a name in the <dose_names> property.
delta: the value of the delta in the comparison for MED, or the quantile for the EDq
alt_name: the short name for the QOI, the default QOIs have specific names, the user added QOIs have names of the form QOI_4_N for the nth target dose QOI
is_default_qoi: “True” or “False”, if the QOI is a default QOI its values should not be changed.
qoi_id: this is a Microsoft unique id, so that the QOI can be referenced from other parameters where the QOI is used.
Example:
<target_dose_qoi>
<target_type>2</target_type>
<tte_qoi_type>1</tte_qoi_type>
<dose_combo_range_type>1</dose_combo_range_type>
<dose_combo_range_arm>Dose 2</dose_combo_range_arm>
<relation>
<is_absolute>True</is_absolute>
<comparison_arm />
<delta>2</delta>
</relation>
<alt_name>QOI_4_1</alt_name>
<is_default_qoi>False</is_default_qoi>
<qoi_id>276464ff-0bfd-4509-b010-564a06e16d85</qoi_id>
</target_dose_qoi>2.2.8 Decision QOI
Note the decision qoi XML tags use a perhap confusing naming scheme the list tag is <decisions_qoi> and each individual entry is <decision_qois>.
The parameters stored in a <decision_qois> structure are:
ref_qoi_id: the id of the post_prob, pred_prob or pvalue QOI to be used
target_dose_qoi: the id of the target dose QOI to use to pick a specific value out of the ref_qoi. Instead of a QOI id, it is also possible to use an “ActualDose”. This has the form “ActualDose- - -” followed by the Dose Name or “Max/Min probability over all doses” or “Max/Min pvalue over all doses”.
alt_name: the short name for the QOI, the default QOIs have specific names, the user added QOIs have names of the form <ref QOI’s short name>_at_<target dose QOI short name or Dose Name>
is_default_qoi: “True” or “False”, if the QOI is a default QOI its values should not be changed.
qoi_id: this is a Microsoft unique id, so that the QOI can be referenced from other parameters where the QOI is used.
Example:
<decision_qois>
<ref_qoi_id>83e1a1cc-dc7b-4ab0-a4f0-f16d1fd20d02</ref_qoi_id>
<target_dose_qoi>975e51d6-8399-4dca-b76c-4bef85fa7b5b</target_dose_qoi>
<alt_name>QOI_1_2_at_Pr(Max)</alt_name>
<is_default_qoi>False</is_default_qoi>
<qoi_id>f8831384-dfc1-4b68-800f-005cf1c1fea2</qoi_id>
</decision_qois>
<decision_qois>
<ref_qoi_id>06288ef6-3979-4eeb-a1b2-2a8e9fc58891</ref_qoi_id>
<target_dose_qoi>ActualDose---Max probability over all doses</target_dose_qoi>
<alt_name>QOI_1_3_at_Max_probability_over_all_doses</alt_name>
<is_default_qoi>False</is_default_qoi>
<qoi_id>ac597934-f638-41df-b770-86b42f8f4bb8</qoi_id>
</decision_qois>2.2.9 Interim Criteria
The parameters stored in an <interim_criteria> are
interim_number: the number of the interim the criteria apply to, the number 32767 indicates the final success/futility criteria
<success_criteria> see below for details
<futility_criteria> see below for details
<graduation_criteria> see below for details
2.2.10 Success / Futility / Graduation Criteria
The parameters stored in a <success_criteria>, <futility_criteria> or <graduation_criteria> are:
is_enabled: True or False
list of one or more <stopping_criterion>, see below for details
is_and_condition: True if conditions are AND-ed together, False if conditions are OR-ed together
min_info_reqd: N – the minimum total amount of information (as defined for interims) required
min_dose_info_reqd:
target dose: The id of a target QOI or “Actual Dose- - -” followed by the Dose name
min_info: N – the minimum information (as defined for interims) on that dose required
2.2.11 Stopping Criterion
The parameters stored in a <stopping_criterion> structure are:
qoi_id: the id of the decision QOI being tested
condition: 3=greater than, 4=less than
threshold: the threshold the QOI is to be compared to
2.2.12 Individual Dose Decision Criterion
The parameters stored in a <individual_dose_decision_ui> structure are:
<dose_name>: The name of the dose for this decision criteria.
<is_keep>: True or False: is the flag a decision type of Keep
<is_and>: True or False: True if is “and” combination with another dose decision
<simple_criterion>: see below for details
2.2.13 Simple Criterion
The parameters stored in a <simple_criterion> structure are:
<qoi_id>: the id of the decision QOI being tested
<condition>: 3=greater than, 4=less than
<threshold>: the threshold the QOI is compared to
2.2.14 Arm Decision Method Type
The parameters stored in a <arm_decision_method> structure are:
<id>: the id of the arm decision method
<name>: the name of the arm decision method
2.2.15 Group Representative UI
The parameters stored in a <group_repesentative_ui structure are:
<is_top>: True or False: true if the group representative is the pre-selected top dose
<is_sum>: True or False: true if the group representative summed across groups
2.2.16 Keep One Group Decision
The parameters stored in a <keep_one_group_decision> structure are:
<qoi_id>: the id of the qoi used.
<condition>: the condition type, 1=Greatest, 2=Least
2.2.17 Dose Combo
The parameters stored in a <doseCombo> structure are:
iddoseY: the row factor strength of the dose combo
iddoseX: the column factor strength of the dose combo
isAllowed: flag indicating whether the dose combo is used in the study
name: the name of the dose combo
level: the effective strength of the dose combo
Example:
<doseCombo>
<iddoseY>0</iddoseY>
<iddoseX>1</iddoseX>
<isAllowed>1</isAllowed>
<name>Dose 1, Default</name>
<level>1</level>
</doseCombo>
2.3 The values
| Tags | Parameter | Number of items | Default Value | Notes |
|---|---|---|---|---|
| < facts | version | 1 | the FACTS GUI version number | FACTS will not open “.facts” files with a higher version number. This test has/will be relaxed to only check the first 2 numbers in the version number. |
| guid | 1 | see general comment on id’s in section 1 | ||
| name | 1 | the name of the “.facts” file | Don’t make it different from the file name, bad things may happen. | |
| host | 1 | |||
| <project-notes | CDATA | 1 | “” | All text from the “Notes” text box |
| </project-notes> | ||||
| < parameterSets | type | 1 | SDBaseParameterSet | used for all FACTS Staged Design parameters |
| <parameterSet | name | 1 | “sd_base” | The staged design base parameter set |
| SetID | 1 | see general comment on id’s in section 1 | ||
| <property> | type | 1 | 1-4 | The Endpoint type: 1= Continuous, 2 = Dichotomous, 3 = TTE, 4 = Multiple Endpoint |
| QOI Parameters | ||||
| <property> | qoi_parameter_set | |||
| <qoi_types> | ||||
| <post_prob_qois> | ||||
| <post_prob_qoi> | PostQ | 2 | This is a list of the Posterior Probability QOIs, see above for details. There are two default Posterior Probability QOIs Pr(PBO) and Pr(CSD) | |
| </post_prob_qois> | ||||
| <pred_prob_qois> | ||||
| PredQ | 1 | This is the list of the Predictive Probability QOIs, see above for details. There is one default Predictive Probability QOI Pr(SS) | ||
| <\pred_prob_qois> | ||||
| <pvalue_qois> | ||||
| PvQ | 0 | This is the list of the p-value QOIs, see above for details. There are no default p-value QOIs. | ||
| <\pvalue_qois> | ||||
| <target_doses_qoi> | ||||
| TQ | 3 | This is the list of the Target Dose QOIs, see above for details. There are three default Target Dose QOIs Pr(Max), Pr(MED), Pr(EDq) | ||
| <\target_doses_qoi> | ||||
| <decision_qoi> | ||||
| DQ | 0 | This is the list of the Decision QOIs, see above for details. There are no default Decision QOIs | ||
| <\decision_qoi> | ||||
| </qoi_types> | ||||
| <property> | is_longitudinal_enabled | 1 | 0 | This flag indicates if longitudinal modeling is enabled. |
| <property> | continuous_endpoint | 1 | 1 | This flag indicates if this is a continuous endpoint |
| <property> | mirror_response_data | 1 | 0 | This flag indicates if response data is mirrored between Stage 1/2 |
| <property> | mirror_long_data | 1 | 0 | Thid flag indicates if longitudinal data is mirrored between Stage 1/2 |
| Treatment Arms Parameters | ||||
| <property> | use_placebo | 1 | 1 | This flag corresponds to the “Include a control treatment arm” option on the Treatment Arms tab. 0=Don’t include, 1=Do include a control treatment arm. |
| <property> | use_ac | 1 | 0 | This flag corresponds to the “Include an additional comparator arm” option on the Treatment Arms tab. 0=Don’t include, 1=Do include an additional comparator arm. |
| <property> | doses_save | D if 2d_model=0; CF if 2d_model=1 |
0, 1 | An item list of the effective dose strengths of the treatment arms. The length of this list defines ‘D’ the total number of doses/arms in the trial. Control, if present, always has a strength of ‘0’. If Control is not present, there should not be a dose/arm with a strength of ‘0’. If the 2D dose model is used, the list has length ‘CF’ and represents the effective dose strengths of the column factors in the trial. |
| <property> | dose_names | D if 2d_model=0; CF if 2d_model=1 |
Control, Dose 1 | An item list of the names of the doses/arms. If the 2D dose model is used, the represents the list of the names of the column factors. |
| <property> | 2d_model | 1 | 0 | Flag indicating whether the 2D dose model should be used. |
| <property> | doses_y | RF | 0, 1 | Only used when the 2D dose model is enabled. This is an item list of the effective row factor strengths. The length of this list defines ‘RF’ the total number of row factors in the trial. |
| <property> | dose_y_names | RF | Control, Factor 1 | Only used when the 2D dose model is enabled. This is an item list of the names of the row factors. |
| <property> | control_dose_combo_y_level | 1 | 0 | Only used when the 2D dose model is enabled. This represent the row factor strength of the selected control dose. This uniquely identifies the selected control dose and its column factor strength will always be zero. |
| <property> | 2d_dose_combos | DC * <doseCombo> | Only used when the 2D dose model is enabled. This represent the full list of dose combinations that can be included in the study. | |
| <property> | doses_xy_save | ADC | 0, 1 | Only used when the 2D dose model is enabled. An item list of the effective dose strengths of the dose combinations included in the study. |
| Visit Info Parameters | ||||
| <property> | study_weeks | V | If is_longitudinal_modeled=0 this is an empty <item /> list, if is_longitudinal_modeled=1, this is a list of items, one per visit, the value of each is the week of the visit. (As listed in the “Schedule of Post-Baseline Visits) | |
| <property> | final_visit | 1 | 1 | This is the week of the last visit, if is_longitudinal_modeled=1 this value should be the same as the last entry in the study_weeks list. |
| <property> | visit_names | V | If is_longitudinal_modeled=0 this is an empty <item /> list, if is_longitudinal_modeled=1, this is a list of items, one per visit, the value of each is the visit name. (As listed in the “Schedule of Post-Baseline Visits” on the Study Info tab) | |
| Study Information Limits Parameters | ||||
| <property> | stage_one_max_subjects | 1 | 0 | Maximum number of subjects for Stage 1 |
| <property> | stage_two_max_subjects_per_arm | 1 | 0 | A flag indicating if stage 2 maximum number of subjects depends on the number of treatment arms as defined in stage 2 allocation TODO: include here, the stage 2 allocation parameter that this depend s on |
| <property> | stage_two_max_subjects | 1 | 0 | Maximum number of subjects for Stage 2 |
| <property> | max_subjects_fulltrial | 1 | 300 | Maximum number of subjects for the full trial. This value must be less than or equal to max subjects for stage 1 and stage 2 combined |
| <property> | max_accrual_time | 1 | 0 | Maximum time in weeks for Stage 1 accrual. |
| <property> | max_doseselection_time | 1 | Not used. | |
| Staged Multiple Endpoint Study Info Parameters | ||||
| <property> | max_subjects_ui | 1 | 300 | Maximum number of subjects in the trial. See max_subjects_fulltrial. |
| <property> | Is_noninferiority | 1 | 0 | Flag for the default QOIs, defining the trial type: 0=Superiority, 1=Non-inferiority. |
| <property> | use_cohorts | 1 | 0 | For Staged ME, only values are: 0=Sequential (continuously) 2=Deterministically |
| <property> | has_variants | 1 | 0 | A flag to indicate if there are variants. |
| <property> | num_variants | 1 | 5 | if has_variants=1 this value is the number of variants |
| <property> | variant_names | Var | “Var 1”, “Var 2”, … | The names of the “variants” to be included in the scenario names to indicate which variant was being simulated. |
| <property> | var_max_overall_subjects | Var | 300, 300, … | The different values of max_subjects_fulltrial to be used in each variant. |
| <property> | var_max_stage1_subjects | Var | “” ““, … | The different values of stage_one_max_subjects to be used in each variant. |
| <property> | var_max_stage2_subjects | Var | “” ““, … | The different values of stage_two_max_subjects to be used in each variant. |
| <property> | var_selection_qoi | 1 | The Target Dose QOI to select the dose. See Target Dose QOI for details. | |
| Allocation Parameters | ||||
| <property> | arm_drop | 1 | 2 | If is_adaptive=1 then this flag (despite its name) actually indicates the type of allocation to be used: 0=legacy adaptive allocation 1=arm dropping 2=fixed allocation 3=adaptive allocation 4=deterministic allocation |
| <property> | alloc_update_type | 1 | 0 | If is_adaptive=1, this flag indicates whether interims are: 0=by time, 1=by information |
| <property> | det_alloc_filename | 1 | “” | If is_adaptive=1 and arm_drop=4 (deterministic allocation), this is the name of the file containing the allocations. |
| <property> | min_alloc_prob | 1 | 0 | if is_adaptive=1 and arm_drop=0 or 3 (adaptive allocation), or if using cohorts and adaptive allocation, this is the floor probability for allocation, if an arm has less than this probability of allocation its probability is set to zero and the probabilities are re-normalized. |
| <property> | arm_drop_limit_save | if is_adaptive=1 and arm_drop=1 (arm dropping), this is the maximum number of arms that can be dropped during the study | ||
| <property> | dropped_arm_alloc | 1 | 0 | if is_adaptive=1 and arm_drop=1 (arm dropping) this is the option of what to do with subject allocation post an arm drop 0=Maintain study size, maintain block size 1=Decrease Study size 2=Maintain study size, reduce block size |
| <property> | prune_protect_low | 1 | 1 | if is_adaptive=1 and arm_drop=1 (arm dropping) this flag (if =1) indicates that arms should be dropped from the lowest, and can only be dropped if all lower arms have been dropped. |
| <property> | prune_protect_high | 1 | 0 | if is_adaptive=1 and arm_drop=1 (arm dropping) this flag (if =1) indicates that arms should be dropped from the highest, and can only be dropped if all higher arms have been dropped. |
| <property> | prune_priority | 1 | 0 | if is_adaptive=1 and arm_drop=1 (arm dropping), there is a limit on how many arms can be dropped, and either both prune_protect_low and prune_protect_high are set, or neither are set, this flags specifies whether to drop the lower or higher dose for preference: 0=prioritize lower 1=prioritize higher |
| <property> | arm_drop_blocking | D | 1, 1, … | these are the static allocation ratios for the different arms. |
| <property> | static_alloc | D | 0, 1, 1, … | if is_adaptive=1 and arm_drop=3 (adaptive allocation) and if the adaptive allocation includes some static allocation this list is the static weights to each arm. |
| <property> | rr_power | 1 | 1 | if is_adaptive=1 and arm_drop=3 (adaptive allocation) this is the power to raise the allocation probabilities by |
| <property> | init_alloc_save | D | 5, 5, … | if is_adaptive=1 and arm_drop=0 or 3 (adaptive allocation), this list of values are the “pre-fist interim allocation ratios” |
| <property> | dose_fix_alloc_save | D | 0, 0, … | if is_adaptive=1 and arm_drop=3 (adaptive allocation), these flags correspond to the “fix Alloc” check boxes: 1=fix alloc to this arm after first interim 0=adaptive allocation to this arm after first interim |
| <property> | dose_fixed_prob_save | D | ““,”“, … | By default blank but with an item per dose. If is_adaptive=1 and arm_drop=0 (legacy adaptive allocation) the items that correspond to the arms for which the “dose_fix_alloc_save” flag has been set, contain the “Post burn-in Alloc Prob” value for the post first interim allocation. |
| <property> | block_size | 1 | 10 | If is_adaptive=1 and arm_drop=0 or arm_drop=3 (adaptive allocation) this is the “post first interim block size”. |
| <property> | block_plac_rate | 1 | 2 | If is_adaptive=1 and arm_drop=0 (legacy adaptive allocation) and Control does not have a fixed probability, then this is the “Control block rate” given the value of block-size (above). |
| <property> | block_alloc_gui | D | ““,”“, … | By default blank but with an item per dose. If is_adaptive=1 and arm_drop=3 (adaptive allocation) the items that correspond to the arms for which the “dose_fix_alloc_save” flag has been set, contain the “Alloc per Block” value for the post first interim allocation. |
| Interims Parameters | ||||
| <property> | update_freq_save | 1 | 2 | If interims are defined by time, this is the update frequency in weeks. |
| <property> | update_subjects | I | 1, … | If interims are defined by the amount of information, these values are the number of subjects/events enrolled / complete / opportunity to complete at each interim |
| <property> | interims_milestones_visits | I | ““,”“, | By default blank but if interims are by the amount of information this is the visit number of the information. |
| <property> | single_burnin_new | 1 | 1 | If interims are by time, this is the amount of information required for the first interim to occur |
| <property> | first_interim_visit | 1 | 1 | If is_adaptive=1, and alloc_info_type=1 or 2 and alloc_update_type=0, (So applies when the interims are by time and information is completed data at specified visit or opportunity to complete at specified visit). Represents the visit number text box when specifying first interim information in the context of interims by time. |
| <property> | continue_follow_success | 1 | 1 | This is the flag for whether to follow-up subjects if the trial has stopped early for success (1=follow-up, 0=don’t follow-up) |
| <property> | continue_follow_futility | 1 | 0 | This is the flag for whether to follow-up subjects if the trial has stopped early for futility (1=follow-up, 0=don’t follow-up) |
| <property> | continue_follow_arm_drop | 1 | 0 | This is the flag for whether to follow-up subjects on an arm that has been dropped (1=follow-up, 0=don’t follow-up) |
| <property> | stop_ia | 1 | 0 | If is_adaptive=1 and alloc_update_type=0, this flag controls whether to “discontinue interim analysis beyond full enrollment”, 0=don’t discontinue, 1=discontinue |
| <property> | continue_followup_grad | 1 | 0 | This is the flag for whether to follow-up subjects if the treatment arm graduates. |
| <property> | s1_complete_data_a_gui | 1 | 1 | This is a flag for whether to perform Stage 1 complete data analysis for output. |
| <property> | dose_selection_censors_output | 1 | 0 | This is a flag for whether to censor Stage 1 patients at dose selection in the output. |
| <property> | delay_type | 1 | 2 | A flag indicating whether to include a delay period after interim, before Stage 2 dose selection, and if included, why type of delay. 0=Wait until a Stage 1 enrollee completes 1=Wait a fixed time 2=No delay |
| <property> | delay_fixed_weeks | 1 | 0 | If delay_type=1, the number of weeks to delay before Stage 2 dose selection. |
| <property> | interim_information_stage1counts | 1 | 0 | This is a flag indicating, in Stage 2, what information an interim is based on: 0=Just Stage2 Data, 1=Stage 2 and included Stage 1 data |
| Staged Multiple Endpoint Stage Parameters | ||||
| <property> | type | 1 | 1 | 1=Continuous endpoint, 2=Dichotomous endpoint |
| <property> | qoi_parameter_set | Not used in Stage Parameter Set. | ||
| <property> | stage_N_data_adaptive_alloca_target_parameter_set | See Adaptive Allocation Target Parameter Set | ||
| <property> | Stage_N_data_arm_dropping_criteria | Not used in Stage Parameter Set | ||
| <property> | stage_N_is_adaptive | 1 | 0 | This is the “Adaptive” option on the Study Info tab. If the flag=1, then the Adaptive box is checked and adaptive options displayed. It needs to be set if adaptive features (interims, allocation) are being used. |
| <property> | stage_N_data_alloc_info_type | 1 | 0 | This is the “Subject Information Defined By:” option: 0=Subjects Enrolled, 1=Complete Data at Specified Visit, 2=Opportunity to Complete at Specified Visit. |
| <property> | stage_N_data_interim_stopping_criteria | This comprises a <interims_criteria> which is a list of <interim_criteria> one per interim for which criteria have been defined plus one for the final criteria. See details above. | ||
| <property> | stage | 1 | 1 | The stage number, 1 or 2 |
| <property> | util_combination | 1 | 1 | A value representing an index into a list of utility combination methods: Two endpoints 0=U1*U2 1=U1+U2 Three endpoints 2=U1*U2*U3 3=U1+U2+U3 4=(U1+U2)*U3 5-U1*(U2+U3) 18=U1+(U2*U3) 19=(U1*U2)+U3 Four endpoints 6=U1*U2*U3*U4 7=U1+U2+U3+U4") 8=(U1+U2+U3)*U4 9=U1*(U2+U3+U4) 10=U1+(U2*U3*U4) 11=(U1*U2*U3)+U4 12=(U1*U2)+(U3*U4) 13=(U1+U2)*(U3+U4) 14=(U1+U2)*U3*U4 15=U1*U2+U3+U4" 16=U1*U2*(U3+U4) 17=U1+U2+U3*U4 20=U1+U2*(U3+U4) 21=U1*(U2+U3)+U4 22=U1*(U2*U3+U4) 23=(U1+U2)*U3+U4 24=(U1+U2*U3)*U4 25=(U1*U2+U3)*U4 |
| <property> | stage_2_data_use | 1 | 0 | A flag indicating how stage 1 data is used in stage2: 0=Ignored, 1=UsedInFull, 2=UsedForStage2Arms, 3=TreatedAsSingleArm, 4=LongModelDelay, 5=NonCompletersUsed |
| <property> | stage2_data_use_visit | 1 | 0 | The maximum number of visits which define a subject as a non-completer. Used when stage2_data_use flag is set to 5. |
| <property> | is_keep_decision | 1 | 1 | A flag to indicate if this is a keep type of decision. Used in Stage 1 Staged Parameter Set only. |
| <property> | all_dose_decisions_and | 1 | 0 | A flag indicting this decision is an “and” combination. Used in Stage 1 Staged Parameter Set only. |
| <property> | group_count | 1 | 1 | Represents the number of groups defined, and is used when graduation_logic_type=1. . Used In Stage 1 Staged Parameter Set only. |
| <property> | keep_control | 1 | 0 | A flag indicating whether to keep the control arm into Stage 2. Used in Stage 1 Staged Parameter Set only. |
| <property> | keep_ac | 1 | 0 | A flag indicating whether to keep the AC arm into Stage 2. Used n Stage 1 Staged Parameter Set only. |
| <property> | operational_delay_time | 1 | 0 | A value in weeks representing the delay between Stage 2 dose selection and start of Stage 2. Used n Stage 1 Staged Parameter Set only. |
| <property> | qoi_use_stage2_arms | 1 | 0 | A flag indicating if Min/Max Decision QOI’s and Target QOI’s are selectable from: 0=all arms, 1=only arms selected for Stage2 |
| Transition Parameters | ||||
| <property> | stage2_data_use | 1 | 0 | A flag indicating how stage 1 data is used in stage2: 0=Ignored, 1=UsedInFull, 2=UsedForStage2Arms, 3=TreatedAsSingleArm, 4=LongModelDelay, 5=NonCompletersUsed |
| <property> | stage2_data_use_visit | 1 | 0 | The maximum number of visits which define a subject as a non-completer. Used when stage2_data_use flag is set to 5. |
| <property> | graduation_logic_type | 1 | 0 | The dose selection method used for graduation logic, default is 0: 0=StandardTransition, 1=RepresentativeArm |
| <property> | individual_dose_decisions_ui | D | A collection of <individual_dose_decision> used only if graduation_logic_type=0 | |
| <individual_dose_decision> | dose_name | 1 | The name of the doses for the individual_dose_decision | |
| is_keep | 1 | True | A flag indicating whether to keep this dose | |
| is_and | 1 | True | A flag indicating if this is an and combination with another <individual_dose_decisions> item. | |
| <simple_criterion> | qoi_id | 1 | The id of the decision QOI being tested | |
| condition | 1 | A value giving the relation (greater, less) of the qoi to a threshold: 3=greater than, 4=less than | ||
| threshold | 1 | 0 | A threshold value to which the qoi will be compared | |
| <property> | keep_best_target_doses | D | A collection of target dose qoi’s, used only if graduation_logic_type=0 | |
| <target_dose_qoi> | target_type | TQ | 2=MED, 3=EDQ | |
| <relation> | is_absolute | 1 | True | True=Absolute, False=Relative to an arm or control |
| comparison_arm | 1 | The arm to compare to | ||
| delta | The value use in the comparison | |||
| <alt_name> | A descriptive name of the qoi | |||
| <is_default_qoi> | Is this a default QOI, True=default, False=created by user | |||
| <qoi_id> | The id of the target dose qoi | |||
| <property> | is_keep_decision | 1 | 1 | A flag to indicate if this is a keep type of decision, used only if graduation_logic_type=0 |
| <property> | min_to_keep | 1 | 1 | Minimum of doses to keep, used in All Dose Decisions |
| <property> | max_to_keep | 1 | 1 | Maximum of doses to keep, used in All Dose Decisions |
| <property> | all_dose_decisions_and | 1 | 0 | A flag indicting this decision is an “and” combination |
| <property> | sort_criterion | 1 | A single Simple Criterion | |
| <simple_criterion> | qoi_id | The id of the decision QOI being tested | ||
| condition | A value giving the relation (greater, less) of the qoi to a threshold: 3=greater than, 4=less than | |||
| threshold | A threshold value to which the qoi will be compared. | |||
| <property> | keep_drop_qoi_decisions | A collection of Simple Criterion used to set up the keep/drop decisions | ||
| <simple_criterion> | qoi_id | 1 | The id of the decision QOI being tested | |
| condition | 1 | A value giving the relation (greater, less) of the qoi to a threshold: 3=greater than, 4=less than | ||
| threshold | 1 | 0 | A threshold value to which the qoi will be compared | |
| <property> | group_a_name | 1 | User defined name for Group A, used when qraduation_logic_type=1 | |
| <property> | group_b_name | 1 | User defined name for Group B, used when qraduation_logic_type=1 | |
| <property> | group_c_name | 1 | User defined name for Group C used when qraduation_logic_type=1 | |
| <property> | group_d_name | 1 | User defined name for Group D, used when qraduation_logic_type=1 | |
| <property> | group_a_dose | 1 | A collection of Doses for Group A, consists of dose name and dose level. Used if graduation_logic_type=1 | |
| <property> | group_b_dose | 1 | A collection of Doses for Group B, consists of dose name and dose level. Used if graduation_logic_type=1 | |
| <property> | group_c_dose | 1 | A collection of Doses for Group C, consists of dose name and dose level. Used if graduation_logic_type=1 | |
| <property> | group_c_dose | 1 | A collection of Doses for Group D, consists of dose name and dose level. Used if graduation_logic_type=1 | |
| <property> | arm_decision_method | 1 | 1 | When graduation_logic_type=1, this is the method used for representative arm decisions. 1=Keep one from each group, 2=Keep group based on representative, 3=Keep one group |
| <property> | group_representative_ui | 1 | 1 | When graduation_logic_type=1, and arm_decision_method=2 or 3, this is the dose which represents the group. This value is either a qoi or a pre-defined value of Pre-selected top dose or Sum across groups. See Group Representative UI |
| <property> | group_count | 1 | 1 | Represents the number of groups defined, and is used when graduation_logic_type=1 |
| <property> | keep_one_from_each_group_decisions | DQ+TQ | The criteria used when deciding to keep one dose from each group. Consists of one or more simple_criterion. Used when graduation_logic_type=1 | |
| <simple_criterion> | qoi_id | 1 | The id of the decision QOI being tested | |
| condition | 1 | 3=greater than, 4=less than | ||
| threshold | 1 | 0 | A threshold value to which the qoi will be compared | |
| <property> | keep_group_on_representative_decisions | DQ+TQ | The criteria used when deciding to a group based on a representative dose. Consists of one or more simple_criterion. Used when graduation_logic_type=????? | |
| <simple_criterion> | qoi_id | 1 | The id of the decision QOI being tested | |
| condition | 1 | A value giving the relation (greater, less) of the qoi to a threshold: 3=greater than, 4=less than | ||
| threshold | 1 | 0 | A threshold value to which the qoi will be compared | |
| <property> | keep_one_group_decision | DQ+TQ | The criteria used to keep one group based on a group representative dose. Contains one or more keep_one_group_decision_criteria | |
<keep_one_group_ decision_criterion> |
qoi_id | The id of the decision qoi | ||
| condition | 1 | A value giving the relation (greater, less) of the qoi to a threshold: 3=greater than, 4=less than | ||
| <property> | combine_arm_decisions_with_and | 1 | 1 | A flag indicting this decision is an “and” combination |
| <property> | keep_control | 1 | 0 | A flag indicating whether to keep the control arm into Stage 2 |
| <property> | keep_ac | 1 | 0 | A flag indicating whether to keep the AC arm into Stage 2 |
| <property | operational_delay_time | 1 | 0 | A value in weeks representing the delay between Stage 2 dose selection and start of Stage 2 |
| <property> | qoi_use_stage2_arms | 1 | 0 | A flag indicating if Min/Max Decision QOI’s and Target QOI’s are selectable from: 0=all arms, 1=only arms selected for Stage2 |
| <property> | stage2_data_use | 1 | 0 | A flag indication how stage 1 data is used in stage2: 0=Ignored, 1=UsedInFull, 2=UsedForStage2Arms, 3=TreatedAsSingleArm, 4=LongModelDelay, 5=NonCompletersUsed |
| <property> | stage2_data_use_visit | 1 | 0 | The maximum number of visits which define a subject as a non-completer. Used when stage2_data_use flag is set to 5. |
| <property> | graduation_logic_type | 1 | 0 | The dose selection method used for graduation logic: 0=StandardTransition, 1=RepresentativeArm |
| <property> | individual_dose_decisions_ui | D | A collection of <individual_dose_decision> | |
| <individual_dose_decision> | dose_name | 1 | The name of the dose | |
| is_keep | 1 | True | A flag indicating whether to keep this dose | |
| is_and | 1 | True | A flag indicating if this is an and combination with another <individual_dose_decisions> | |
| <simple_criterion> | qoi_id | 1 | The id of the decision QOI being tested | |
| condition | 1 | 3=greater than, 4=less than | ||
| threshold | 1 | 0 | The threshold to be compared to | |
| <property> | keep_best_target_doses | D | A collection of target dose qoi’s | |
| <target_dose_qoi> | target_type | TQ | 2=MED, 3=EDQ | |
| <relation> | is_absolute | 1 | True | True=Absolute, False=Relative to an arm or control |
| comparison_arm | 1 | The arm to compare to | ||
| delta | The value use in the comparison | |||
| <alt_name> | A descriptive name of the qoi | |||
| <is_default_qoi> | Is this a default QOI, True=default, False=created by user | |||
| <qoi_id> | The id of the target dose qoi | |||
| <property> | is_keep_decision | 1 | 1 | A flag to indicate if this is a keep type of decision |
| <property> | min_to_keep | 1 | 1 | Minimum of doses to keep, used in All Dose Decisions |
| <property> | max_to_keep | 1 | 1 | Maximum of doses to keep, used in All Dose Decisions |
| <property> | all_dose_decisions_and | 1 | 0 | A flag indicting this decision is an “and” combination |
| <property> | sort_criterion | 1 | A single Simple Criterion | |
| <simple_criterion> | qoi_id | The id of the decision QOI being tested | ||
| condition | 3=greater than, 4=less than | |||
| threshold | The threshold to be compared to | |||
| <property> | keep_drop_qoi_decisions | A collection of Simple Criterion used to set up the keep/drop decisions | ||
| <simple_criterion> | qoi_id | 1 | The id of the decision QOI being tested | |
| condition | 1 | 3=greater than, 4=less than | ||
| threshold | 1 | 0 | The threshold to be compared to | |
| <property> | group_a_name | 1 | User defined name for Group A, used when qraduation_logic_type=1 | |
| <property> | group_b_name | 1 | User defined name for Group B, used when qraduation_logic_type=1 | |
| <property> | group_c_name | 1 | User defined name for Group C used when qraduation_logic_type=1 | |
| <property> | group_d_name | 1 | User defined name for Group D, used when qraduation_logic_type=1 | |
| <property> | group_a_dose | 1 | A collection of Doses for Group A, consists of dose name and dose level. Used if graduation_logic_type=1 | |
| <property> | group_b_dose | 1 | A collection of Doses for Group B, consists of dose name and dose level. Used if graduation_logic_type=1 | |
| <property> | group_c_dose | 1 | A collection of Doses for Group C, consists of dose name and dose level. Used if graduation_logic_type=1 | |
| <property> | group_c_dose | 1 | A collection of Doses for Group D, consists of dose name and dose level. Used if graduation_logic_type=1 | |
| <property> | arm_decision_method | 1 | 1 | When graduation_logic_type=1, this is the method used for representative arm decisions. 1=Keep one from each group, 2=Keep group based on representative, 3=Keep one group |
| <property> | group_representative_ui | 1 | 1 | When graduation_logic_type=1, and arm_decision_method=2 or 3, this is the dose which represents the group. This value is either a qoi or a pre-defined value of Pre-selected top dose or Sum across groups. See Group Representative UI |
| <property> | group_count | 1 | 1 | Represents the number of groups defined, and is used when graduation_logic_type=1 |
| <property> | keep_one_from_each_group_decisions | DQ+TQ | The criteria used when deciding to keep one dose from each group. Consists of one or more simple_criterion. Used when graduation_logic_type=1 | |
| <simple_criterion> | qoi_id | 1 | The id of the decision QOI being tested | |
| condition | 1 | 3=greater than, 4=less than | ||
| threshold | 1 | 0 | The threshold to be compared to | |
| <property> | keep_group_on_representative_decisions | DQ+TQ | The criteria used when deciding to a group based on a representative dose. Consists of one or more simple_criterion. Used when graduation_logic_type=????? | |
| <simple_criterion> | qoi_id | 1 | The id of the decision QOI being tested | |
| condition | 1 | 3=greater than, 4=less than | ||
| threshold | 1 | 0 | The threshold to be compared to | |
| <property> | keep_one_group_decision | DQ+TQ | The criteria used to keep one group based on a group representative dose. Contains one or more keep_one_group_decision_criteria | |
<keep_one_group_ decision_criterion> |
qoi_id | The id of the decision qoi | ||
| condition | 1 | 3=greater than, 4=less than | ||
| <property> | combine_arm_decisions_with_and | 1 | 1 | A flag indicting this decision is an “and” combination |
| <property> | keep_control | 1 | 0 | A flag indicating whether to keep the control arm into Stage 2 |
| <property> | keep_ac | 1 | 0 | A flag indicating whether to keep the AC arm into Stage 2 |
| <property | operational_delay_time | 1 | 0 | A value in weeks representing the delay between Stage 2 dose selection and start of Stage 2 |
| <property> | qoi_use_stage2_arms | 1 | 0 | A flag indicating if Min/Max Decision QOI’s and Target QOI’s are selectable from: 0=all arms, 1=only arms selected for Stage2 |
| Stage Two Allocation Parameters | ||||
| <property> | alloc_matrix_data_gui | If fixed allocation, this is the per treatment arm allocation for stage 2 | ||
| <property> | ac_allocation | If fixed allocation, this is the allocation to active comparator for stage 2 | ||
| <property> | block_size_vector | If fixed allocation, this is the block size for treatment arms for stage 2 | ||
| <property> | max_subjects_per_arm_vector_ui | This is the maximum number of subjects per arm for stage 2 | ||
| <property> | control_allocation_gui | If fixed allocation, this is the allocation to control for stage 2 | ||
| <property> | burnin_matrix_data_gu | If adaptive allocation, this is the burn-in allocation per treatment arm for stage 2 | ||
| <property> | burnin_control_allocation_gui | If adaptive allocation, this is the burn-in allocation to the control arm for stage 2 | ||
| <property> | burnin_ac_allocation | If adaptive allocation, this is the burn-in allocation to active comparator for stage 2 | ||
| Endpoint Parameters | ||||
| <parameterSets> | type | EndpointParameterSet | The collection of endpoint parameter sets, 1 for each endpoint. | |
| <parameterSet> | name | 4 | Primary Endpoint, Endpoint 1-2 | |
| <property> | type | 1 | 1 | The endpoint type. 1=Continuous, 2=Dichotomous |
| <property> | qoi_parameter_set | |||
| <qoi_types> | ||||
| <post_prob_qois> | ||||
| <post_prob_qoi> | PostQ | 2 | This is a list of the Posterior Probability QOIs, see above for details. There are two default Posterior Probability QOIs Pr(PBO) and Pr(CSD) | |
| </post_prob_qois> | ||||
| <pred_prob_qois> | ||||
| PredQ | 1 | This is the list of the Predictive Probability QOIs, see above for details. There is one default Predictive Probability QOI Pr(SS) | ||
| <\pred_prob_qois> | ||||
| <pvalue_qois> | ||||
| PvQ | 0 | This is the list of the p-value QOIs, see above for details. There are no default p-value QOIs. | ||
| <\pvalue_qois> | ||||
| <target_doses_qoi> | ||||
| TQ | 3 | This is the list of the Target Dose QOIs, see above for details. There are three default Target Dose QOIs Pr(Max), Pr(MED), Pr(EDq) | ||
| <\target_doses_qoi> | ||||
| <decision_qoi> | ||||
| DQ | 0 | This is the list of the Decision QOIs, see above for details. There are no default Decision QOIs | ||
| <\decision_qoi> | ||||
| </qoi_types> | ||||
| </property | ||||
| <property> | endpoint_name | 1 | The endpoint name, auto supplied, modifiable. | |
| <property> | continuous_endpoint | 1 | 0 | A flag indicating if this is a continuous endpoint. 0=false, 1=true |
| <property> | use_longitudinal | 1 | 1 | Flag – 0=Don’t use longitudinal modelling, 1=Use longitudinal modelling [Not sure why there is both this and the next parameter – it seems they are either both 0 or both 1] |
| <property> | use_stable_state | 1 | 0 | A flag indicating to use Restricted Markov model. 0=false, 1=true. If the endpoint is set to use Longitudinal Modeling and Use restricted Markov model is checked, use_stable_state = 1, else = 0 |
| <property> | high_is_good | 1 | 1 | This flag corresponds to the “Higher/Lower response is subject improvement” selection on the Study->Endpoints tab. 0=Lower, 1=Higher. If this is a Dichotomous endpoint design it corresponds to the equivalent “Response indicates positive/negative outcome” selection on the Study->Endpoints tab. 0=negative, 1=positive |
| <property> | is_noninferiority | 1 | 0 | Flag for the default QOIs, defining the endpoint trial type: 0=Superiority, 1=Non-inferiority. |
| <property> | csd | 1 | 1/0.1 | This the CSD value for the default QOIs on the Quantities of Interest tab. The default value is 1 for a Continuous endpoint, 0.1 for a Dichotomous endpoint. |
| <property> | target_is_rel | 1 | 1 | A flag to indicate if the target dose is relative or absolute comparison to control |
| <property> | include_baseline | 1 | 0 | An option for the Continuous endpoint. A flag to indicate whether including simulation of baseline: 0=Don’t simulate baseline, 1= Simulate baseline |
| <property> | is_change_from_baseline | 1 | 1 | An option for the Continuous endpoint, if simulating baseline. A flag to indicate whether response analyzed is absolute or change from baseline: 0=Change from baseline, 1= Final endpoint value |
| <property> | final_visit | 1 | 1 | This is the week of the last visit, if is_longitudinal_modeled=1 this value should be the same as the last entry in the study_weeks list. |
| <property> | visit_weeks | V | If use_longitudinal=1, this is the per visit flags to indicate if the visit_week item is included in longitudinal modeling. | |
| <property> | final_stable_success | 1 | 1 | An option for the Dichotomous endpoint. If use_special_long=1 AND special_long_feature=1, this flag indicates whether a subject whose final state is “stable” will be counted as a Responder or Failure, 0=Failure, 1=Responder |
| <property> | util_is_rel | 1 | 1 | ? |
| <property> | utility_intervals | Collection of parms for the utility endpoint | ||
| <utility> | ||||
| lower | 1 | -Infinty | ? | |
| upper | 1 | Infinity | ? | |
| alpha | 1 | 0 | ? | |
| beta | 1 | 1 | ? | |
| gamma | 1 | 0 | ? | |
| delta | 1 | 0 | ? | |
| Epsilon | 1 | 0 | ? | |
| phi | 1 | 0 | ? | |
| psi | 1 | 1 | ? | |
| </utility> | ||||
| </property> | ||||
| <property> | output_freq_locf | 1 | 1 | A flag that indicates whether a frequentist file is to be output with frequentist analysis results using LOCF for missing data. 1=output, 0=don’t output |
| <property> | output_freq_bocf | 1 | 0 | (Applies to continuous only) A flag that indicates whether a frequentist file is to be output with frequentist analysis results using BOCF for missing data. 1=output, 0=don’t output |
| <property> | output_freq_ignore | 1 | 0 | A flag that indicates whether a frequentist file is to be output with frequentist analysis results ignoring missing data. 1=output, 0=don’t output |
| <property> | output_freq_fail | 1 | 0 | (Applies to dichotomous only) A flag that indicates whether a frequentist file is to be output with frequentist analysis results using fail for missing data. 1=output, 0=don’t output |
| <property> | do_dunnetts | 1 | 0 | A flag that indicates whether any frequentist file output is to include a Dunnett’s adjustment analysis. 1=include, 0=don’t include |
| <property> | endpoint_number | 1 | 1 | The endpoint number, auto assigned. |
| <property> | stage_number | 1 | 1 | The stage number, auto assigned. |
| </parameterSet> | ||||
| </parameterSets> | ||||
| Utility Parameters | ||||
| <parameterSets> | type | EndpointParameterSet | The collection of endpoint parameter sets, 1 for each endpoint. | |
| <parameterSet> | name | 4 | Primary Endpoint, Endpoint 1-2 | |
| <property> | endpoint_number | 1 | 1 | The endpoint number to which this utility is related to |
| <property> | util_is_rel | 1 | 1 | A flag indicating if this utility response is parametrized relative to control arm |
| <property> | utility_intervals | Collection of values for the utility endpoint | ||
| <utility> | ||||
| lower | 1 | The upper value for the interval | ||
| upper | 1 | The lower value for the interval | ||
| alpha | 1 | 0 | The alpha value for the utility function | |
| beta | 1 | 1 | The beta value for the utility function | |
| gamma | 1 | 0 | The gamma value for the utility function | |
| delta | 1 | 0 | The alpha value for the utility function | |
| Epsilon | 1 | 0 | The delta value for the utility function | |
| phi | 1 | 0 | The phi value for the utility function | |
| psi | 1 | 1 | The psi value for the utility function | |
| </utility> | ||||
| </property> | ||||
| </parameterSet> | ||||
| </parameterSets> | ||||
| Staged Multiple Endpoint Frequentists Parameters | ||||
| <property> | one_sided_alpha | 1 | 0.025 | The one-sided alpha for the standard frequentist analysis |
| <property> | output_freq_locf | 1 | 1 | A flag that indicates whether a frequentist file is to be output with frequentist analysis results using LOCF for missing data. 1=output, 0=don’t output |
| <property> | output_freq_bocf | 1 | 0 | (Applies to continuous only) A flag that indicates whether a frequentist file is to be output with frequentist analysis results using BOCF for missing data. 1=output, 0=don’t output |
| <property> | output_freq_ignore | 1 | 0 | A flag that indicates whether a frequentist file is to be output with frequentist analysis results ignoring missing data. 1=output, 0=don’t output |
| <property> | output_freq_fail | 1 | 0 | (Applies to dichotomous only) A flag that indicates whether a frequentist file is to be output with frequentist analysis results using fail for missing data. 1=output, 0=don’t output |
| <property> | do_dunnetts | 1 | 0 | A flag that indicates whether any frequentist file output is to include a Dunnett’s adjustment analysis. 1=include, 0=don’t include |
| <property> | contrast_coeff_save | D | 0, 1, 2, … | The contrast coefficients for the various arms to be used in a trend test in the standard frequentist analysis |
| <property> | freq_noninferiority _delta | 1 | 0 | The frequentist super-superiority / non-inferiority delta value for p-value QOI’s and/or standard frequentist analysis |
| <property> | freq_noninferiority _qoi_only | 1 | 1 | A flag for whether the non-inferiority delta is to be used for QOI’s only and not for regular frequentist analysis |
| <property> | use_fisher_exact | 1 | 0 | A flag for whether to use a Fisher exact test for p-value calculation, including current trial predictive probability QOIs. Only applies for a dichotomous endpoint. |
| <property> | endpoint_number | 1 | 1 | The endpoint number this <frequentists_settings> parameter set is associated with |
| <property> | stage_number | 1 | 1 | The stage number this <frequentists_settings> parameter set is associated with |
| Response Model Parameters | ||||
| <property> | efficacy_model | 1 | 1 | A flag indicating the type of the dose response model: 1=Simple NDLM 4=3-Parameter Logistic 5=Hierarchical Logistic 6=Sigmoidal 7=Monotonic NDLM 8=3-Param Exp. Logistic (Dichotomous only) 9=2-Parameter Logistic (Dichotomous only) 10=Independent Dose Model 12=U-shaped model 13=Plateau Model 14=Second Order NDLM 15=Simple Hierarchical 16=Simple Linear 17=Simple Hierarchical Linear 21=2D continuous factorial model (2D dose model only) 22=2D discrete factorial model (2D dose model only) 23=2D NDLM model (2D dose model only) |
| <property> | prior_sigma | <gamma> | 10, 1 | A parameter for the Continuous endpoint. Contains the values for the inverse-gamma prior for ’sigma” the estimate of the SD of the endpoint. |
| <property> | ndlm_tau | <gamma> | 10, 1 | Contains the values for the inverse-gamma prior for the smoothing parameter (tau) of any NDLM dose response model (τ2) |
| <property> | hier_lambda | <gamma> | 1,1 | Contains the values for the inverse-gamma prior for the hierarchical part of the Hierarchical Logistic dose response model (α42) |
| <property> | logit_min | <meansdpair> | 0, 10 | Contains the values for the Normal prior for the minimum response (at control) of any logistic dose response model - (α1) in the 3-parameter logistic, hierarchical logistic and sigmoidal dose response models. |
| <property> | logit_max | <meansdpair> | 0, 10 | Contains the values for the Normal prior for the maximum response of any logistic dose response model - (α2). |
| <property> | logit_ed50 | <meansdpair> | 5, 5 | Contains the values for the Normal prior for the ED50 dose strength of any logistic dose response model - (α3). |
| <property> | logit_slope | <meansdpair> | 1, 5 | Contains the values for the Normal prior for the slope parameter of the 4-parameter logistic dose response model - (α4) in the sigmoidal dose response model |
| <property> | ushaped_theta | <meansdpair> | 0, 1 | Contains the values for the Normal prior for the response at control in the plateau or u-shaped dose response model - (θ0) |
| <property> | ushaped_delta | <meansdpair> | 1, 1 | Contains the values for the Normal-positive prior for delta, the maximum response (relative to control) in the plateau or u-shaped dose response model - (δ) |
| <property> | ushaped_pmin | <meansdpair> | 2, 3 | Contains the values for the Normal-positive prior for the estimate of the dose strength at the start of the maximum response in the plateau or u-shaped dose response model - (pmin) |
| <property> | ushaped_pwidth | <meansdpair> | 3, 3 | Contains the values for the Normal-positive prior for the width of the dose range where the maximum response is observed in the u-shaped dose response model – (pwidth) |
| <property> | ushaped_alpha | <meansdpair> | 1, 0.5 | Contains the values for the Log-Normal prior for the slope in the initial rising section of the model (0<υd<pmin) in the plateau or u-shaped dose response model - (δ) |
| <property> | ushaped_wwidth | <meansdpair> | 3, 3 | Contains the values for the Normal-positive prior for the width of the dose range where the maximum response returns to the control respose in the u-shaped dose response model – (wwidth) |
| <property> | ushaped_beta | <meansdpair> | 1, 0.5 | Contains the values for the Log-Normal prior for the response at control in the u-shaped dose response model – (β) |
| <property> | monotonic_increasing | 1 | 2 | Flag for monotonic models indicating whether the response is 2=increasing or 1=decreasing with dose strength. (Applies to monotonic NDLM, U-shaped and plateau models) [The logistic models can be either increasing or decreasing – depending on the estimate of α2] |
| <property> | model_control_separately | 1 | 0 | Flag indicating if the estimate of the treatment effect of the control arm is modelled separately from the rest of the dose response. 0=included in the overall dose response model, 1=modelled separately |
| <property> | historical_data_save | 1 | 0 | This flag is a duplicate of historical_data_bac, it is recommended that it should be set to mirror that value. |
| <property> | historical_data_ac | 1 | 0 | This flag indicates if the estimation of the treatment effect of the active comparator arm is using a hierarchical prior. 0=not using a hierarchical prior, 1=using a hierarchical prior |
| <property> | historical_data_bac | 1 | 0 | This flag indicates if the estimation of the treatment effect if the control arm is using a hierarchical prior. 0=not using a hierarchical prior, 1=using a hierarchical prior. Note that the legacy parameter historical_data_save should be set to match the value of this flag. Also note that to use a hierarchical prior the control arm must be being “modelled separately” from the study arms, so model_placebo_separately must be =1. |
| <property> | control_prior | <meansdpair> | 0, 10 | If (model_placebo_separately=1 OR no_model_single_prior=0) AND historical_data_bac=0, then these values are the Normal prior for the estimate for the treatment effect of the Control arm. |
| <property> | comparator_prior | <meansdpaoir> | 0, 10 | If an active comparator arm has been included and historical_data_ac=0, then these values are the Normal prior for the estimate for the treatment effect of the Active Comparator arm. |
| <property> | bac_study_names | CS | “” | If historical_data_bac=1, this is the list of study names for the hierarchical prior for control |
| <property> | ac_study_names | CS | “” | If historical_data_bac=1, this is the list of study names for the hierarchical prior for comparator |
| <property> | ss_bac_study | CS | “” | If historical_data_bac=1, this is the list of sample sizes for each study for the prior for control |
| <property> | ss_baac_study | ACS | “” | if historical_data_ac=1, this is the list of sample sizes for each study for the prior for AC |
| <property> | cont_resp_bac_study | CS | “” | If historical_data_bac=1, this is the list of mean responses for each study for the prior for control |
| <property> | cont_resp_ac_study | ACS | “” | if historical_data_ac=1, this is the list of mean responses for each study for the prior for AC |
| <property> | dich_resp_bac_study | CS | “” | If historical_data_bac=1, this is the list of response rates for each study for the prior for control |
| <property> | dich_resp_ac_study | ACS | “” | if historical_data_ac=1, this is the list of response rates for each study for the prior for AC |
| <property> | sd_bac_study | CS | “” | If historical_data_bac=1, this is the list of SDs of the response for each study for the prior for control |
| <property> | sd_baac_study | ACS | “” | if historical_data_ac=1, this is the list of SDs of the response rate for each study for the prior for AC |
| <property> | hyperprior_mean | 1 | 0 | If historical_data_bac=1, this value is the mean of the prior of the mean of the hierarchical model of the response on the control arm (μ0) |
| <property> | hyperprior_mean_ac | 1 | 0 | if historical_data_ac=1, this value is the mean of the prior of the mean of the hierarchical model of the response on the active comparator arm (μAC) |
| <property> | hyperprior_sd | 1 | 1 | If historical_data_bac=1, this value is the SD of the prior of the mean of the hyperprior of the hierarchical model for the response on the control arm (μ0) |
| <property> | hyperprior_sd_ac | 1 | 1 | if historical_data_ac=1, this value is the SD of the prior of the mean of the hyperprior of the hierarchical model for the response on the control arm (μAC) |
| <property> | hyperprior_tau | <gamma> | 1,1 | Contains the values for the inverse-gamma prior for the variance of the hyperprior of a hierarchical prior on Control (τ02). |
| <property> | hyperprior_tau_ac | <gamma> | 1,1 | Contains the values for the inverse-gamma prior for the variance of the hyperprior of a hierarchical prior on Active Comparator (τAC2). |
| <property> | init_dose_prior | <meansdpair> | 0, 10 | |
| <property> | no_model_mean | D | 0, 0, ... | If no_model_single_prior=0, then these values are the mean of the Normal prior for the treatment effect for each study arm. |
| <property> | no_model_sd | D | 10, 10, … | If no_model_single_prior=0, then these values are the SD of the Normal prior for the treatment effect for each study arm. |
| <property> | use_beta_binomial_single_prior | 1 | 1 | Flag indicating whether (if using the independent beta binomial dose model) the same prior applies to all study doses 1=Same prior, 0=Prior per dose (and model control separately) If use_beta_binomial_single_prior =1 then the values for the prior for all doses are stored in beta_binomial_single_prior If use_beta_binomial_single_prior =0 then the prior values per study arm are stored in beta_binomial_alpha and beta_binomial_beta |
| <property> | beta_binomial_single_prior | <betadist> | 2, 2 | If use_beta_binomial_single_prior=1, this is the beta distribution prior for the independent Beta binomial dose model |
| <property> | beta_binomial_alpha | D | 2, 2, ... | If use_beta_binomial_single_prior =0, then these values are the alpha parameter values of the Beta distribution prior for the treatment effect for each study arm. |
| <property> | beta_binomial_beta | D | 2, 2, … | If use_beta_binomial_single_prior =0, then these values are the beta parameter values of the Beta distribution prior for the treatment effect for each study arm. |
| <property> | baseline_in_model | 1 | 0 | Flag indicating (if baseline is being simulated) if the treatment effect includes a baseline adjustment. 0=No baseline adjustment, 1=Use baseline adjusted model |
| <property> | baseline_beta | <meansdpair> | 0, 1 | The vales for the Normal prior for the estimate of the baseline coefficient |
| <property> | dropout_imputation | 1 | 0 | Flag indicating how missing data due to dropout will be handled, 0=Bayesian multiple imputation post baseline, 1=Baseline carried forward, 2=Last observation carried forward |
| <property> | ndlm2_first_diff | <meansdpair> | 0, 10 | If the dose response model is the second order NDLM then this parameter contains the values for the Normal prior of the “first dose difference” (θ1-θ0) of that model |
| <property> | one_sided_alpha | 1 | 0.025 | The one-sided alpha for the standard frequentist analysis |
| <property> | contrasts | D | 0, 1, 2, … | The contrast coefficients for the various arms to be used in a trend test in the standard frequentist analysis |
| <property> | endpoint_number | 1 | The endpoint number for this parameter set. | |
| <property> | stage_number | 1 | 1 | Not used in Core Multiple Endpoint. |
| <property> | contin_2d_interaction | 1 | 1 | Flag indicating whether the 2D continuous factorial dose response model includes an interaction term |
| <property> | contin_2d_alpha | 4*<meansdpair> | 0, 10 | The values for the Normal prior of the α_0, α_1, α_3 and α_3 parameters of the 2D continuous factorial dose response model |
| <property> | discrete_2d_interaction | 1 | 1 | Not available in FACTS 6.3.0. Flag indicating whether the 2D discrete factorial dose response model includes an interaction term |
| <property> | discrete_2d_a | <meansdpair> | 0, 10 | The values for the Normal prior of the α parameter of the 2D discrete factorial dose response model |
| <property> | discrete_2d_b | (CF -1)*<meansdpair> | 0, 10 | The values for the Normal prior of the β_2,…, β_CF parameters of the 2D discrete factorial dose response model |
| <property> | discrete_2d_c | (RF -1)*<meansdpair> | 0, 10 | The values for the Normal prior of the γ_2,…, γ_RF parameters of the 2D discrete factorial dose response model |
| <property> | discrete_2d_delta_mu | 2*<meansdpair> | 0, 10 | Not available in FACTS 6.3.0. The values for the Normal hyperprior of the interaction terms means µ_δ^h_k (k = β, γ) parameters of the 2D discrete factorial dose response model |
| <property> | discrete_2d_delta_sigmasq | 2*<gamma> | 1, 2 | Not available in FACTS 6.3.0. The values for the inverse gamma hyperprior of the interaction terms variances σ_δ^h_k (k = β, γ) parameters of the 2D discrete factorial dose response model |
| <property> | ndlm_2d_usefixedratio | 1 | 1 | Flag indicating whether a fixed smoothing ratio should be used to compute φ (from τ) in the 2D NDLM model |
| <property> | ndlm_2d_ratiok | 1 | 1 | Only used when ndlm_2d_usefixedratio=1. This represents the value of the fixed smoothing ratio used to compute φ (from τ) in the 2D NDLM model |
| <property> | ndlm_2d_phi | <gamma> | 10, 1 | Only used when ndlm_2d_usefixedratio=0. This represents the values of the inverse gamma prior of φ in the 2D NDLM model |
| <property> | ndlm_2d_priordose_strength | 1 | 0 | This represents the dose strength of the initial dose specified in the 2D NDLM model. Note that its Normal prior values are specified in init_dose_prior |
| Longitudinal Model Parameters | ||||
| <property> | long_model_type | 1 | 3 | Flag that indicates the type of the longitudinal model. For a Continuous endpoint these are: 1=linear regression, 3=last observation carried forward, 4=time course hierarchical, 5=kernel density, 8=ITP For a Dichotomous endpoint: 2=Beta binomial 3=last observation carried forward, 7=Logistic regression |
| <property> | long_model_spec | 1 | 0 | This flag indicates the number of longitudinal model instances 0=Single model for all arms 1=Model control separately 2=Model comparator separately 3=Model control and comparator separately 4=Model all arms separately |
| <property> | long_prior_spec | 1 | 0 | This flag indicates the number of separate priors for the longitudinal model parameters 0=Same prior across all model instances and visits 1=Specify per visit, across all model instances 2=Specify priors per model instance and visit |
| <property> | long_alpha_mu | 1 | 0 | When long_prior_spec=0, contains the values for the mean of the Normal prior for the intercept of the linear regression, models (αt) to be used for all models This is the value used when the same prior is used across all model instances and visits. |
| <property> | long_alpha_mu_matrix | LM*V | 0, 0, … | When long_prior_spec=1 or 2, contains the values for the mean of the Normal prior for the intercept of the linear regression, models (αt) with different values for each visit, and possibly each model. See the description of LM*V parameters before this table. |
| <property> | long_alpha_stdv | When long_prior_spec=0, contains the values for the SD of the Normal prior for the intercept of the linear regression, models (αt) to be used for all models This is the value used when the same prior is used across all model instances and visits. |
||
| <property> | long_alpha_sd_matrix | When long_prior_spec=1 or 2, contains the values for the SD of the Normal prior for the intercept of the linear regression, models (αt) with different values for each visit, and possibly each model. See the description of LM*V parameters before this table. |
||
| <property> | long_beta_mu | When long_prior_spec=0, contains the values for the mean of the Normal prior for the slope of the linear regression, models (βt) to be used for all models This is the value used when the same prior is used across all model instances and visits. |
||
| <property> | long_beta_mu_matrix | When long_prior_spec=1 or 2, contains the values for the mean of the Normal prior for the slope of the linear regression, models (βt) with different values for each visit, and possibly each model. See the description of LM*V parameters before this table. |
||
| <property> | long_beta_stdv | When long_prior_spec=0, contains the values for the SD of the Normal prior for the slope of the linear regression, models (βt) to be used for all models This is the value used when the same prior is used across all model instances and visits. |
||
| <property> | long_beta_sd_matrix | When long_prior_spec=1 or 2, contains the values for the SD of the Normal prior for the slope of the linear regression, models (βt) with different values for each visit, and possibly each model. See the description of LM*V parameters before this table. |
||
| <property> | tch_tau | <gamma>*LM=The number of longitudinal models | 0.5, 1 | Contains the values for the inverse-gamma prior for the variance of the inter-subject error term (δi) of the Time Course Hierarchical longitudinal model (τ2). The number of prior values depends on the number of longitudinal models, even if the same priors are used across all models. In which case the first values are used for all models. |
| <property> | tch_alpha | <meansdpair>*LM*V | 0, 2, … | Whether long_prior_spec=0 or 2, these are the priors for the αt parameter in the Time Course Hierarchical longitudinal model. αt is the log of the fraction of the final response seen at visit t. Unlike other LMs there always separate prior values for each visit. If long_prior_spec=0 only the values for the first model in the list are set. See the description of LM*V parameters before this table. |
| <property> | itp_k | <meansdpair>*LM | 1,1 | Contains the values for the Normal prior for the ITP shape parameter (k) of the ITP longitudinal model. The number of prior values depends on the number of longitudinal models, even if the same priors are used across all models. In which case the first values are used for all models. |
| <property> | itp_tau | <gamma>*LM | 1,1 | Contains the values for the inverse-gamma prior for the variance of the inter-subject error term (si) (subject specific random effect) of the ITP longitudinal model (τ2). The number of prior values depends on the number of longitudinal models, even if the same priors are used across all models. In which case the first values are used for all models. |
| <property> | itp_omega | <meansdpair>*LM | 1,1 | Contains the values for the Normal prior for the mean final response of the ITP longitudinal model. The number of prior values depends on the number of longitudinal models, even if the same priors are used across all models. In which case the first values are used for all models. |
| <property> | long_gamma | <gamma> | 0.5, 1 | Contains the values for the inverse-gamma prior for the variance of the residual error of the linear regression, time course hierarchical and ITP longitudinal models (λ2). This is the value used when the same prior is used across all model instances and visits. |
| <property> | long_lambda_list | <gamma>*LM | 0.5, 1 | Contains the values for the inverse-gamma prior for the variance of the residual error of the time course hierarchical and ITP longitudinal models (λ2). This is the value used when a different prior is used across all model instances. These models do not have separate estimates of this parameter per visit, unlike the linear regression… |
| <property> | long_lambda_matrix | <gamma>*LM*V | 0.5, 1 | Contains the values for the inverse-gamma prior for the variance of the residual error of the linear regression longitudinal model (λ2). See the description of LM*V parameters before this table. |
| <property> | kernel_bandwidth | LM | 2 | This is the value of the kernel bandwidth parameter for the Kernel Density longitudinal model. The value is a list of items - one for each LM model. If long_prior_spec=0, and there are multiple models, just the first value in the list is used, even though entries for the other models are present. |
| <property> | kernel_min_subj | LM | 6 | This is the value of the “minimum number of subjects before using” parameter for the Kernel Density longitudinal model. The value is a list of items - one for each LM model. If long_prior_spec=0, and there are multiple models, just the first value in the list is used, even though entries for the other models are present. |
| <property> | kernel_sd | LM | 2 | This is the value of the “standard deviation of the prior mean” parameter for the Kernel Density longitudinal model. The value is a list of items - one for each LM model. If long_prior_spec=0, and there are multiple models, just the first value in the list is used, even though entries for the other models are present. |
| <property> | bb_alpha_mu_1 | 1 | 1 | When long_prior_spec=0, this is the single prior for the αμ1 parameter for the Beta Binomial prior for the probability that a subject with a response at a visit will have a response at the final visit – for all models and all visits. |
| <property> | bb_alpha_mu_1_full | LM*V | 1, 1, … | When long_prior_spec=1 or 2, these are the priors for the αμ1 parameter for the Beta Binomial prior for the probability that a subject with a response at a visit will have a response at the final visit across the different visits or models and visits. See the description of LM*V parameters before this table. |
| <property> | bb_beta_mu_1 | 1 | 1 | When long_prior_spec=0, this is the single prior for the βμ1 parameter for the Beta Binomial prior for the probability that a subject with a response at a visit will have a response at the final visit – for all models and all visits. |
| <property> | bb_beta_mu_1_full | LM*V | 1, 1, … | When long_prior_spec=1 or 2, these are the priors for the βμ1 parameter for the Beta Binomial prior for the probability that a subject with a response at a visit will have a response at the final visit across the different visits or models and visits. See the description of LM*V parameters before this table. |
| <property> | bb_alpha_mu_0 | 1 | 1 | When long_prior_spec=0, this is the single prior for the αμ0 parameter for the Beta Binomial prior for the probability that a subject without a response at a visit will have a response at the final visit – for all models and all visits. |
| <property> | bb_alpha_mu_0_full | LM*v | 1, 1, … | When long_prior_spec=1 or 2, these are the priors for the αμ0 parameter for the Beta Binomial prior for the probability that a subject without a response at a visit will have a response at the final visit across the different visits or models and visits. See the description of LM*V parameters before this table. |
| <property> | bb_beta_mu_0 | 1 | 1 | When long_prior_spec=0, this is the single prior for the βμ0 parameter for the Beta Binomial prior for the probability that a subject without a response at a visit will have a response at the final visit – for all models and all visits. |
| <property> | bb_beta_mu_0_full | LM*V | 1, 1, … | When long_prior_spec=1 or 2, these are the priors for theβμ0 parameter for the Beta Binomial prior for the probability that a subject without a response at a visit will have a response at the final visit across the different visits or models and visits. See the description of LM*V parameters before this table. |
| <property> | long_rmm_alpha_0 | LM*V | 1, 1, … | Whether long_prior_spec=0 or 2, these are the priors for the α0 parameter for the Dirichlet prior in the Restricted Markov Model, for the probability that stable subject will be a fail at the final visit across the different visits. Unlike other LMs there always separate prior values for each visit. If long_prior_spec=0 only the values for the first model in the list are set. See the description of LM*V parameters before this table. |
| <property> | long_rmm_alpha_S | LM*V | 1, 1, … | Whether long_prior_spec=0 or 2, these are the priors for the α0 parameter for the Dirichlet prior in the Restricted Markov Model, for the probability that stable subject will be stable at the final visit across the different visits. Unlike other LMs there always separate prior values for each visit. If long_prior_spec=0 only the values for the first model in the list are set. See the description of LM*V parameters before this table |
| <property> | long_rmm_alpha_1 | LM*V | 1, 1, … | Whether long_prior_spec=0 or 2, these are the priors for the α0 parameter for the Dirichlet prior in the Restricted Markov Model, for the probability that stable subject will be respnder at the final visit across the different visits. Unlike other LMs there always separate prior values for each visit. If long_prior_spec=0 only the values for the first model in the list are set. See the description of LM*V parameters before this table |
| <property> | lr_mu_1 | 1 | 0.5 | When long_prior_spec=0, this is the single prior for the μ1 parameter for the Normal prior in the Logistic Regression longitudinal model, for the logit of the probability that a subject with a response at a visit will have a response at the final visit – for all models and all visits. |
| <property> | lr_mu_1_matrix | LM*V | 0.5, 0.5 | When long_prior_spec=1 or 2, these are the priors for the μ1 parameters for the Normal prior in the Logistic Regression longitudinal model, for the logit of the probability that a subject with a response at a visit will have a response at the final visit across the different visits or models and visits. See the description of LM*V parameters before this table. |
| <property> | lr_mu_0 | 1 | 0.5 | When long_prior_spec=0, this is the single prior for the μ0 parameter for the Normal prior in the Logistic Regression longitudinal model, for the logit of the probability that a subject without a response at a visit will have a response at the final visit – for all models and all visits. |
| <property> | lr_mu_0_matrix | LM*V | 0.5, 0.5 | When long_prior_spec=1 or 2, these are the priors for the μ0 parameters for the Normal prior in the Logistic Regression longitudinal model, for the logit of the probability that a subject without a response at a visit will have a response at the final visit across the different visits or models and visits. See the description of LM*V parameters before this table. |
| <property> | lr_sd_1 | 1 | 0.5 | When long_prior_spec=0, this is the single prior for the σ1 parameter for the Normal prior in the Logistic Regression longitudinal model, for the logit of the probability that a subject with a response at a visit will have a response at the final visit – for all models and all visits. |
| <property> | lr_sd_1_matrix | LM*V | 0.5, 0.5 | When long_prior_spec=1 or 2, these are the priors for the σ1 parameters for the Normal prior in the Logistic Regression longitudinal model, for the logit of the probability that a subject with a response at a visit will have a response at the final visit across the different visits or models and visits. See the description of LM*V parameters before this table. |
| <property> | lr_sd_0 | 1 | 0.5 | When long_prior_spec=0, this is the single prior for the σ0 parameter for the Normal prior in the Logistic Regression longitudinal model, for the logit of the probability that a subject without a response at a visit will have a response at the final visit – for all models and all visits. |
| <property> | lr_sd_0_matrix | LM*V | 0.5, 0.5 | When long_prior_spec=1 or 2, these are the priors for the σ0 parameters for the Normal prior in the Logistic Regression longitudinal model, for the logit of the probability that a subject without a response at a visit will have a response at the final visit across the different visits or models and visits. See the description of LM*V parameters before this table. |
| <property> | endpoint_number | 1 | 1 | The endpoint number for this set |
| <property> | stage_number | 1 | 1 | Not used for Core Multiple Endpoint |
| Dropout Parameters | ||||
| <parameterSets | type | 1 | DropoutParameterSet | The collection of dropout profiles |
| <parameterSet | name | <Profile name> | The name of the dropout profile | |
| SetID | see general comment on id’s in section 1 | |||
| <property> | is_stage_two_data | 1 | 0 | flag used to indicate that (if=1) the profile is to be used in stage 2 of a staged design. |
| <property> | enable_per_visist_dropouts | 1 | 0 | flag used to indicate the type of dropout profile 0: no dropouts 1: dropouts per dose 2: dropouts per dose per visit Note that the next two properties are always output by FACTS even if not required. |
| <property> | dropout_rate | D | 0.05 | An items list with a drop out rates per treatment arm |
| <property> | visit_drop_out_rate | D*V | 0 | An items list with an items list per visit, that (inner) items list is a list of dropout rates at that visit per treatment arm |
| <property> | use_common_dropout_profile | 1 | 1 | If Staged Design, a flag indicating if Stage 1 and Stage 2 should use a common Dropout profile. |
| </parameterSet> | ||||
| </parameterSets> | ||||
| Baseline Parameters | ||||
| <parameterSets | type | 1 | BaselineParameterSet | The collection of baseline profiles (only included if simulating baseline is enabled) |
| <parameterSet | name | 1 | <profile names> | The name of the baseline profile |
| <property> | baseline_vsr_mean | 1 | 0 | The baseline mean |
| <property> | baseline_vsr_sd | 1 | 1 | The SD of the baseline |
| <property> | constrain_lower | 1 | 0 | Flag to indicate if the lower bound of the baseline is constrained 0: Not constrained, 1: Constrained |
| <property> | constrain_upper | 1 | 0 | Flag to indicate if the lower bound of the baseline is constrained 0: Not constrained, 1: Constrained |
| <property> | lower_bound | 1 | 0 | The lower bound on the baseline score if it is constrained. |
| <property> | upper_bound | 1 | 0 | The upper bound on the baseline score if it is constrained |
| <property> | baseline_adj | 1 | 0 | Flag to indicate if the subjects final response is to be adjusted for baseline, 0: No adjustment, 1: Adjusted |
| <property> | baseline_beta_adj | 1 | 1 | The value for “Beta” the coefficient of the adjustment |
| <property> | baseline_vsr_center | 1 | 0 | The centering value for the adjustment |
| <property> | baselin_vsr_scale | 1 | 1 | The scaling value for the adjustment |
| <property> | endpoint_number | 1 | 1 | The endpoint number for this parameter set. |
| <property> | stage_number | 1 | 1 | If Staged Design, the stage number for this parameter set |
| </parameterSets> | ||||
| Efficacy Parameters | ||||
| <parameterSets | type | 1 | EfficacyParameterSet | The collection of dose response VSR profiles |
| <parameterSet | name | <Profile name> | The name of the Virtual Subject Response profile | |
| <property> | true_endpoint_response | D=number of doses | 0 | This is the simulated mean change from baseline for each dose. |
| <property> | enable_per_dose_sigma | 1 | 0 | This flag indicates whether 0: there is just one value for sigma used for all doses or 1: a value for sigma has been supplied for each dose |
| <property> | true_sigma_save | 1 | 10 | The single value for sigma (used when enable_per_dose_sigma=0) |
| <property> | true_sigma_per_dose_save | D | 10 | A list of values for sigma, one per dose (use when enable_per_dose_sigma=1) |
| <property> | true_response_rate | D | 0.5 | Used in the Dichotomous endpoint response simulation method |
| <property> | external_vsr_mean | 1 | 0 | The “Use external mean vsr file” flag. 0: don’t use (use the parameters here) 1: use the external file |
| <property> | external_mean_vsr_file | 1 | The name of the “mvsr” file, blank if not used. This is the full path name to the fiie. | |
| <property> | should_succeed | D | 0 | The “should succeed” flag from the GUI: 0=not set, 1=set. Use by the GUI in reporting the results, not used by the simulator. |
| <property> | endpoint_number | 1 | 1 | The endpoint number for this parameter set. |
| <property> | stage_number | 1 | 1 | If Staged Design, the stage number for this parameter set |
| </parameterSets> | ||||
| Longitudinal Parameters | ||||
| <parameterSets | type | 1 | LongitudinalParameterSet | The collection of longitudinal VSR profiles (only included if longitudinal simulation enabled). The set of properties included is the union of all the property types for all the different types of longitudinal simulation method (across all endpoint types). |
| <parameterSet | name | <Profile name> | The name of the dropout profile | |
| SetID | see general comment on id’s in section 1 | |||
| <property> | sim_method_save | 1 | 1 | The simulation method: 1: Correlation 2: Hierarchical 6: ITP 8: Hierarchical MMRM |
| <property> | true_rho | V=the number of visits | 0, 08, 0.8, … | Only used in the Correlation method. This is the correlation in response at visit relative to the previous visit (hence 0 for the first visit) |
| <property> | true_long_frac | V | 1,1,1 | Used in the Correlation and Hierarchical methods. This is the fraction of the final response seen at the visit (ft). |
| <property> | hier_nu | 1 | 0.33 | Used in the Hierarchical methods. This is the fraction of the overall variance due to inter-subject variance (ω) [the greek ‘nu’ is no longer used in the GUI for this as it conflicts with the use of ‘nu’ for dose strength, the XML parameter name never changed though] |
| <property> | is_long_true_sigma_per_dose | 1 | 0 | Used in the Correlation and Hierarchical methods 0: fraction of final variance is specified per visit, 1: fraction of final variance is specified per dose per visit. |
| <property> | long_true_sigma | V | 1,1,1 | Used in the Correlation and Hierarchical methods if is_long_true_sigma_per_dose=0. This is the fraction of the final variance observed per visit, the final value must =1 as it is the fraction at the final visit. |
| <property> | long_true_sigma_per_dose | V*D | 1,1,1, …. | Used in the Correlation and Hierarchical methods if is_long_true_sigma_per_dose=1. This is the list of fraction of the final variance observed per dose, per visit. This is a list per visit, of lists of the fractions per dose. The final list values must =1 as the fraction for all doses at the final visit must be 1. |
| <property> | true_retain | V | 0.9 | Used in the Dichotomous endpoint longitudinal method. |
| <property> | true_enter | V | 0.2 | Used in the Dichotomous endpoint longitudinal method. |
| <property> | sim_itp_k | D | 1,1… | Used in the ITP method. This is the ITP shape parameter per dose. |
| <property> | sim_itp_omega | D | 1,1… | Used in the ITP method this is the fraction of the total variance that is inter-subject variance [i.e. similar to ω in the hierarchical methods], per dose. |
| <property> | endpoint_number | 1 | 1 | The endpoint number for this parameter set. |
| <property> | stage_number | 1 | 1 | If Staged Design, the stage number for this parameter set |
| </parameterSets> | ||||
| Composite Response Parameters | ||||
| <parameterSets | type | 1 | CompositeResponseSet | The collection of composite response. |
| <parameterSet> | name | <ProfileName> | ||
| <property> | primary_baseline | 1 | “” | The SetId of the Baseline Parameter Set for the primary baseline of this composite response. |
| <property> | primary_response | 1 | “” | The SetId of the Efficacy Parameter Set for the primary response of this composite response. |
| <property> | primary_longitudinal | 1 | “” | The SetId of the Longitudinal Parameter Set for the primary longitudinal composite response. |
| <property> | secondary_baseline | 1 | “” | The SetId of the Baseline Parameter Set for the secondary baseline of this composite response. |
| <property> | secondary _response | 1 | “” | The SetId of the Efficacy Parameter Set for the secondary response of this composite response. |
| <property> | secondary _longitudinal | 1 | “” | The SetId of the Longitudinal Parameter Set for the secondary longitudinal composite response. |
| <property> | tertiary_baseline | 1 | “” | The SetId of the Baseline Parameter Set for the tertiary baseline of this composite response. |
| <property> | tertiary_response | 1 | “” | The SetId of the Efficacy Parameter Set for the tertiary response of this composite response. |
| <property> | tertiary_longitudinal | 1 | “” | The SetId of the Longitudinal Parameter Set for the tertiary longitudinal composite response. |
| <property> | quaternary_baseline | 1 | “” | The SetId of the Baseline Parameter Set for the quaternarybaseline of this composite response. |
| <property> | quaternary_response | 1 | “” | The SetId of the Efficacy Parameter Set for the quaternaryresponse of this composite response. |
| <property> | quaternary_longitudinal | 1 | “” | The SetId of the Longitudinal Parameter Set for the quaternarylongitudinal composite response. |
| <property> | should_succeed | D | 0 | The “should succeed” flag from the GUI: 0=not set, 1=set. Use by the GUI in reporting the results, not used by the simulator. |
| </parameterSet> | ||||
| </parameterSets | ||||
| Recruitment Parameters | ||||
| <parameterSets | type | 1 | RecruitmentParameterSet | The collection of accrual profiles |
| <parameterSet | name | <Profile name> | The name of the dropout profile | |
| SetID | see general comment on id’s in section 1 | |||
| <property> | is_stage_two_data | 1 | 0 | Only used in staged designs: 0 = this is the accrual profile for the first stage (or both stages) |
| <property> | location_names | R=number of profiles in accrual profile | <Region name> | The name of the regions – with an entry (<item>) for each region. All properties in this profile should have the same number of items. |
| <property> | recruit_rates | R | 5 | The peak mean accrual rate per region |
| <property> | start_dates | R | 0 | The start date (in weeks from the start of the trial) of accrual per region |
| <property> | ramp_up | R | 0 | A flag per region indicating if there is an initial ramp up stage (0= no ramp up, 1=ramp up) |
| <property> | peak_start | R | Include but leave blank if no ramp up. A per region value of the end (in weeks from the start of the trial) of ramp up in that region | |
| <property> | ramp_down | R | 0 | A flag per region indicating if there is a ramp down stage (0= no ramp down, 1=ramp down) |
| <property> | start_day | R | Include but leave blank if no ramp down. A per region value of the start (in weeks – despite the parameter name - from the start of the trial) of ramp down in that region | |
| <property> | last_day | R | Include but leave blank if no ramp down. A per region value of the end (in weeks – despite the parameter name - from the start of the trial) of ramp down in that region | |
| <property> | common_accrual_profile | R | 1 | Only used in staged designs: 1 = indicates that there is one accrual profile used for both stages. |
| </parameterSets> | ||||
| <simulation-settings | ||||
| <property> | num_sims | 1 | 5 | The number of simulations to run |
| <property> | burn | 1 | 1000 | The number of iterations of the MCMC loop that are initially ignored before sampling starts |
| <property> | length | 1 | 2500 | The number of iterations of the MCMC loop used to estimate the parameters |
| <property> | rv_seed_1 | 1 | 3500 | The first part (32 bits) of the random number seed for all simulations. |
| <property> | rv_seed_2 | 1 | 1 | The second part (32 bits) of the random number seed for all simulations. |
| <property> | num_weeks_out | 1 | 100 | The number of simulations which will have their interim analysis results written out to file. |
| <property> | num_freq_weeks_out | 1 | 1 | The number of simulations which will have their interim frequentist analysis results written out to file. |
| <property> | num_subj_out | 1 | 1 | The number of simulations which will have their subject data written out to file. |
| <property> | num_mcmc_files | 1 | 0 | The number of simulations which will have their MCMC samples written out to file. |
| <property> | mcmc_thin | 1 | 1 | This is the factor by which the MCMC samples are thinned *when writing out the MCMC file*, it does not thin the samples used within FACTS. There is no option to thin the sampling within FACTS. |
| <property> | mcmc_per_imputation | 1 | 1 | The number of MCMC samples between each imputation of missing data using the longitudinal model |
| <property> | is_local | 1 | 1 | |
| </simulation-settings> | ||||
| </facts> | ||||