Intelligencia GraphQL API Reference

Welcome to the Intelligencia API. This API provides data regarding clinical programs which are in active development (otherwise called ongoing programs). Specifically, Intelligencia’s data is structured around the following main entities:

  • Clinical Program: A program (also known as clinical pipeline or drug pipeline) is the clinical development of a drug (or a set of drugs in case of combination therapies) by a pharmaceutical company (alone or in collaboration with other partners) for an indication. A program consists of a set of clinical trials with the ultimate goal of approval for marketing.
  • Program Characteristics: Program characteristics are all the defining characteristics of a clinical program beyond its drug(s)/intervention(s), indication, primary sponsor and adjuvant value. It includes information that describes the program’s patient population, for example line of treatment, stage of disease, molecular characterization, age, sex, smoking status and previous treatments.
  • Clinical Trial: A clinical trial is a research study conducted in humans which assesses a treatment or a medical intervention. A trial may be linked to more than one program for example when a) a trial tests the same drugs/interventions in multiple indications (Basket Trial), b) a trial tests different drugs/interventions in the same indication (Umbrella Trial), c) a trial that tests multiple subsets of the same indication (e.g subsets of breast cancer), or multiple patient groups under the same indication. Note that a trial can be both basket and umbrella, namely it can test multiple different drug therapies in multiple indications.
  • Clinical Trial Arm: A trial arm describes the different groupings of the trial’s patients based on the administered drug(s)/intervention(s). Any difference in drug(s)/intervention(s) administered to a patient, be it a different drug, a different combination of drugs, a different dose or even a different treatment schedule constitutes a separate arm.
  • Clinical Trial Cohort: A cohort separates patients in further subgroupings based on common patient traits e.g. common genetic makeup, age, smoking status etc. Cohorts are used by the investigators for the analysis of the trial’s outcomes. A trial’s cohort may belong to one or multiple trial arms.
  • Clinical Trial Endpoint: An endpoint measures the outcomes of the trial. Endpoints are usually part of the clinical trial design. When a trial starts the information that is usually available includes the number and type/subtype of endpoints. Later in the trial, when outcomes are published e.g. in papers, conferences, company announcements, then clinical endpoints are filled with the relevant results (aka outcomes).
  • Program Drivers: Drivers are the features that the ML algorithms take into account to predict the program’s PTRS. Intelligencia’s PTRS drivers belong to feature families such as Biology, Clinical Trial Outcomes, Clinical Trial Design, Regulatory, Company Characteristics.
  • Benchmarks: Intelligencia’s benchmarks consist of the historical approval rate and the number of ongoing and historical (approved, failed) programs (or program-trial pairs) for key Biology, Regulatory and Trial Design dimensions. Benchmarks are provided per indication and phase where applicable and are based on data spanning across the last 20 years of clinical development.
  • Drug: Intelligencia’s drug library contains expertly curated information about drugs that are currently in ongoing clinical programs including drug synonyms, mechanisms of action, targets, modalities, genes, biological pathways, protein classes.
  • Target: Intelligencia’s library of targets along with information about their synonyms, related organism and target type (eg. protein, gene). Drugs can have multiple targets.
  • Indication Hierarchy: Intelligencia’s hand-curated therapeutic area-indication ontology for all covered therapeutic areas and indications.
  • Modality Hierarchy: Modality hierarchy is an expertly curated multilevel ontology of medical modalities, such as cell therapies, antibodies etc. Each drug may be linked to more than one modality.
  • Phase Hierarchy: It describes the hierarchy of clinical trial phases.
Contact

API Support Team
[email protected]

API Endpoints
Production:
https://api.intelligencia.ai/v1/graphql
Intelligencia's Graph
  • Check graph explorer here
Intelligencia's API Quick Start Guide
  • Check quick start guide here
Intelligencia's Postman Collection
  • Check our Postman collection here
Intelligencia's API Explorer
  • Check our API Explorer here
API limits

All entities have a default row limit of 250. API consumers are expected to make use of the offset operator for batch retrieval of large datasets.

Queries

api_benchmarks

Fetch benchmarks.
Returns [api_benchmarks!]!

Name Description
distinct_on - [String!] distinct select on columns
limit - Int limit the number of rows returned
offset - Int skip the first n rows. Use only with order_by
order_by - [String!] sort the rows by one or more columns
where - bool_comparison_exp filter the rows returned

Example

Query
query api_benchmarks($distinct_on: [String!], $limit: Int, $offset: Int, $order_by: [String!], $where: bool_comparison_exp) {
  api_benchmarks(distinct_on: $distinct_on, limit: $limit, offset: $offset, order_by: $order_by, where: $where) {
    approved
    benchmark_category
    benchmark_id
    benchmark_type
    failed
    historical
    historical_approval_rate
    indication
    metric
    metric_value
    ongoing
    phase
  }
}
Variables
{
  "distinct_on": [
    "column_name"
  ],
  "limit": 987,
  "offset": 123,
  "order_by": [
    "column_name"
  ],
  "where": bool_comparison_exp
}
Response 
{
  "data": {
    "api_benchmarks": [
      {
        "approved": 10,
        "benchmark_category": "Biology",
        "benchmark_id": 987,
        "benchmark_type": "trial",
        "failed": 146,
        "historical": 156,
        "historical_approval_rate": "6%",
        "indication": "Lymphoma",
        "metric": "therapy_type",
        "metric_value": "Combination",
        "ongoing": 187,
        "phase": "Phase 2"
      }
    ]
  }
}

api_benchmarks_aggregate

Fetch aggregated results for benchmarks.

count, min, max can be used here.

Returns [api_benchmarks_aggregate!]!

Name Description
distinct_on - [String!] distinct select on columns
limit - Int limit the number of rows returned
offset - Int skip the first n rows. Use only with order_by
order_by - [String!] sort the rows by one or more columns
where - bool_comparison_exp filter the rows returned

Example

Query
query api_benchmarks_aggregate($distinct_on: [String!], $limit: Int, $offset: Int, $order_by: [String!], $where: bool_comparison_exp) {
  api_benchmarks_aggregate(distinct_on: $distinct_on, limit: $limit, offset: $offset, order_by: $order_by, where: $where) {
    aggregate {
      ...aggregate_fieldsFragment
    }
    nodes {
      ...api_benchmarksFragment
    }
  }
}
Variables
{
  "distinct_on": [
    "column_name"
  ],
  "limit": 123,
  "offset": 123,
  "order_by": [
    "column_name"
  ],
  "where": bool_comparison_exp
}
Response 
{
  "data": {
    "api_benchmarks_aggregate": [
      {
        "aggregate": aggregate_fields,
        "nodes": [api_benchmarks]
      }
    ]
  }
}

api_drug

Fetch drugs and any of their related types (api_program_drugs, api_trial_drugs)
Returns [api_drug!]!

Name Description
distinct_on - [String!] distinct select on columns
limit - Int limit the number of rows returned
offset - Int skip the first n rows. Use only with order_by
order_by - [String!] sort the rows by one or more columns
where - bool_comparison_exp filter the rows returned

Example

Query
query api_drug($distinct_on: [String!], $limit: Int, $offset: Int, $order_by: [String!], $where: bool_comparison_exp) {
  api_drug(distinct_on: $distinct_on, limit: $limit, offset: $offset, order_by: $order_by, where: $where) {
    action_types
    biological_pathways
    drug_id
    genes
    mechanism_of_action
    modalities
    preferred_name
    program_drugs {
      ...api_program_drugFragment
    }
    protein_class
    synonyms
    targets
    treatment_types
    trial_drugs {
      ...api_trial_drugFragment
    }
  }
}
Variables
{
  "distinct_on": [
    "column_name"
  ],
  "limit": 987,
  "offset": 987,
  "order_by": [
    "column_name"
  ],
  "where": bool_comparison_exp
}
Response 
{
  "data": {
    "api_drug": [
      {
        "action_types": [{"name": Antagonist}],
        "biological_pathways": [
          {"name": Adaptive Immune System},
          {"name": Costimulation by the CD28 family},
          {"name": Immune System},
          {"name": PD-1 signaling}
        ],
        "drug_id": 987,
        "genes": [{"name": PDCD1}],
        "mechanism_of_action": [
          {
            "name": Programmed cell death protein 1 Antagonist
          }
        ],
        "modalities": [{"name": Monoclonal antibody}],
        "preferred_name": "Nivolumab",
        "program_drugs": [api_program_drug],
        "protein_class": [{"name": Receptor}],
        "synonyms": [
          {"name": BMS 936558},
          {"name": BMS-936558},
          {"name": BMS-936558-01},
          {"name": BMS-986298},
          {"name": MDX 1106},
          {"name": MDX-1106},
          {"name": NIVO},
          {"name": Nivolumab},
          {"name": Nivolumab BMS},
          {"name": ONO-0123},
          {"name": ONO 4538},
          {"name": ONO-4538},
          {"name": Opdivo}
        ],
        "targets": [
          {"name": Programmed cell death protein 1}
        ],
        "treatment_types": [{"name": Immunotherapy}],
        "trial_drugs": [api_trial_drug]
      }
    ]
  }
}

api_drug_aggregate

Fetch aggregated results for drugs.

count, min, max can be used here.

Returns [api_drug_aggregate!]!

Name Description
distinct_on - [String!] distinct select on columns
limit - Int limit the number of rows returned
offset - Int skip the first n rows. Use only with order_by
order_by - [String!] sort the rows by one or more columns
where - bool_comparison_exp filter the rows returned

Example

Query
query api_drug_aggregate($distinct_on: [String!], $limit: Int, $offset: Int, $order_by: [String!], $where: bool_comparison_exp) {
  api_drug_aggregate(distinct_on: $distinct_on, limit: $limit, offset: $offset, order_by: $order_by, where: $where) {
    aggregate {
      ...aggregate_fieldsFragment
    }
    nodes {
      ...api_drugFragment
    }
  }
}
Variables
{
  "distinct_on": [
    "column_name"
  ],
  "limit": 123,
  "offset": 987,
  "order_by": [
    "column_name"
  ],
  "where": bool_comparison_exp
}
Response 
{
  "data": {
    "api_drug_aggregate": [
      {
        "aggregate": aggregate_fields,
        "nodes": [api_drug]
      }
    ]
  }
}

api_indication_hierarchy

Returns [api_indication_hierarchy!]!

Name Description
distinct_on - [String!] distinct select on columns
limit - Int limit the number of rows returned
offset - Int skip the first n rows. Use only with order_by
order_by - [String!] sort the rows by one or more columns
where - bool_comparison_exp filter the rows returned

Example

Query
query api_indication_hierarchy($distinct_on: [String!], $limit: Int, $offset: Int, $order_by: [String!], $where: bool_comparison_exp) {
  api_indication_hierarchy(distinct_on: $distinct_on, limit: $limit, offset: $offset, order_by: $order_by, where: $where) {
    child_level
    distance
    indication_child
    indication_parent
    parent_level
  }
}
Variables
{
  "distinct_on": [
    "column_name"
  ],
  "limit": 987,
  "offset": 987,
  "order_by": [
    "column_name"
  ],
  "where": bool_comparison_exp
}
Response 
{
  "data": {
    "api_indication_hierarchy": [
      {
        "child_level": 987,
        "distance": 123,
        "indication_child": "Breast carcinoma",
        "indication_parent": "Breast cancer",
        "parent_level": 123
      }
    ]
  }
}

api_modality_hierarchy

Returns [api_modality_hierarchy!]!

Name Description
distinct_on - [String!] distinct select on columns
limit - Int limit the number of rows returned
offset - Int skip the first n rows. Use only with order_by
order_by - [String!] sort the rows by one or more columns
where - bool_comparison_exp filter the rows returned

Example

Query
query api_modality_hierarchy($distinct_on: [String!], $limit: Int, $offset: Int, $order_by: [String!], $where: bool_comparison_exp) {
  api_modality_hierarchy(distinct_on: $distinct_on, limit: $limit, offset: $offset, order_by: $order_by, where: $where) {
    child_level
    distance
    modality_child
    modality_parent
    parent_level
  }
}
Variables
{
  "distinct_on": [
    "column_name"
  ],
  "limit": 123,
  "offset": 987,
  "order_by": [
    "column_name"
  ],
  "where": bool_comparison_exp
}
Response 
{
  "data": {
    "api_modality_hierarchy": [
      {
        "child_level": 123,
        "distance": 987,
        "modality_child": "Trastuzumab",
        "modality_parent": "abc123",
        "parent_level": 987
      }
    ]
  }
}

api_phase_hierarchy

Returns [api_phase_hierarchy!]!

Name Description
distinct_on - [String!] distinct select on columns
limit - Int limit the number of rows returned
offset - Int skip the first n rows. Use only with order_by
order_by - [String!] sort the rows by one or more columns
where - bool_comparison_exp filter the rows returned

Example

Query
query api_phase_hierarchy($distinct_on: [String!], $limit: Int, $offset: Int, $order_by: [String!], $where: bool_comparison_exp) {
  api_phase_hierarchy(distinct_on: $distinct_on, limit: $limit, offset: $offset, order_by: $order_by, where: $where) {
    phase_child
    phase_parent
  }
}
Variables
{
  "distinct_on": [
    "column_name"
  ],
  "limit": 123,
  "offset": 987,
  "order_by": [
    "column_name"
  ],
  "where": bool_comparison_exp
}
Response 
{
  "data": {
    "api_phase_hierarchy": [
      {
        "phase_child": "Phase 1b",
        "phase_parent": "Phase 1"
      }
    ]
  }
}

api_program

Fetch programs and any of their related types (api_program_drugs, api_program_latest_trial, etc.)
Returns [api_program!]!

Name Description
distinct_on - [String!] distinct select on columns
limit - Int limit the number of rows returned
offset - Int skip the first n rows. Use only with order_by
order_by - [String!] sort the rows by one or more columns
where - bool_comparison_exp filter the rows returned

Example

Query
query api_program($distinct_on: [String!], $limit: Int, $offset: Int, $order_by: [String!], $where: bool_comparison_exp) {
  api_program(distinct_on: $distinct_on, limit: $limit, offset: $offset, order_by: $order_by, where: $where) {
    adjuvant
    administration_mode
    end_date
    indication
    intervention
    latest_trial
    latest_trial_with_outcomes
    link_to_platform
    primary_sponsor
    program_characteristics {
      ...api_program_characteristicsFragment
    }
    program_drivers {
      ...api_program_driversFragment
    }
    program_drugs {
      ...api_program_drugFragment
    }
    program_id
    program_latest_trial {
      ...api_trialFragment
    }
    program_latest_trial_with_outcomes {
      ...api_trialFragment
    }
    program_ptrs {
      ...api_program_ptrsFragment
    }
    program_trial_arm_cohorts {
      ...api_program_trial_arm_cohortFragment
    }
    program_trial_pems {
      ...api_program_trial_pemFragment
    }
    regulatory_pathways
    start_date
    status
    therapeutic_area
    therapy_type
  }
}
Variables
{
  "distinct_on": [
    "column_name"
  ],
  "limit": 987,
  "offset": 123,
  "order_by": [
    "column_name"
  ],
  "where": bool_comparison_exp
}
Response 
{
  "data": {
    "api_program": [
      {
        "adjuvant": [{"name": Neoadjuvant}],
        "administration_mode": Injectable/Intravenous,
        "end_date": 2023-12-08,
        "indication": Leukemia, myeloid, acute,
        "intervention": [{"name": Radio}],
        "latest_trial": 789,
        "latest_trial_with_outcomes": 789,
        "link_to_platform": "https://insight.intelligencia.ai/insight/portfolio-optimizer/program-assessment/deep-dive#programId=986",
        "primary_sponsor": Jasper Therapeutics,
        "program_characteristics": [
          api_program_characteristics
        ],
        "program_drivers": [api_program_drivers],
        "program_drugs": [api_program_drug],
        "program_id": 986,
        "program_latest_trial": api_trial,
        "program_latest_trial_with_outcomes": api_trial,
        "program_ptrs": api_program_ptrs,
        "program_trial_arm_cohorts": [
          api_program_trial_arm_cohort
        ],
        "program_trial_pems": [api_program_trial_pem],
        "regulatory_pathways": [{"name": Orphan drug}],
        "start_date": date,
        "status": "Ongoing",
        "therapeutic_area": [{"name": Oncology}],
        "therapy_type": Combination
      }
    ]
  }
}

api_program_aggregate

Fetch aggregated results for programs.

count, min, max can be used here.

Returns [api_program_aggregate!]!

Name Description
distinct_on - [String!] distinct select on columns
limit - Int limit the number of rows returned
offset - Int skip the first n rows. Use only with order_by
order_by - [String!] sort the rows by one or more columns
where - bool_comparison_exp filter the rows returned

Example

Query
query api_program_aggregate($distinct_on: [String!], $limit: Int, $offset: Int, $order_by: [String!], $where: bool_comparison_exp) {
  api_program_aggregate(distinct_on: $distinct_on, limit: $limit, offset: $offset, order_by: $order_by, where: $where) {
    aggregate {
      ...aggregate_fieldsFragment
    }
    nodes {
      ...api_programFragment
    }
  }
}
Variables
{
  "distinct_on": [
    "column_name"
  ],
  "limit": 987,
  "offset": 987,
  "order_by": [
    "column_name"
  ],
  "where": bool_comparison_exp
}
Response 
{
  "data": {
    "api_program_aggregate": [
      {
        "aggregate": aggregate_fields,
        "nodes": [api_program]
      }
    ]
  }
}

api_trial

Fetch trials and any of their related types (api_arm, api_trial_drugs, etc.)
Returns [api_trial!]!

Name Description
distinct_on - [String!] distinct select on columns
limit - Int limit the number of rows returned
offset - Int skip the first n rows. Use only with order_by
order_by - [String!] sort the rows by one or more columns
where - bool_comparison_exp filter the rows returned

Example

Query
query api_trial($distinct_on: [String!], $limit: Int, $offset: Int, $order_by: [String!], $where: bool_comparison_exp) {
  api_trial(distinct_on: $distinct_on, limit: $limit, offset: $offset, order_by: $order_by, where: $where) {
    acronym
    additional_sponsors
    adverse_events_criteria
    allocation
    arms {
      ...api_armFragment
    }
    basket
    biomarker_subgroup_analysis
    end_date
    endpoints {
      ...api_endpointFragment
    }
    enrollment
    indications
    intervention_model
    last_update_posted_date
    masking
    num_of_countries
    num_of_sites
    phase
    primary_completion_date
    primary_sponsors
    program_trial_arm_cohorts {
      ...api_program_trial_arm_cohortFragment
    }
    results_first_posted_date
    results_first_submitted_date
    safety_acceptable
    source_id
    source_link
    start_date
    study_first_submitted_date
    termination_date
    title
    trial_characteristics {
      ...api_trial_characteristicsFragment
    }
    trial_drugs {
      ...api_trial_drugFragment
    }
    trial_id
    trial_status
    umbrella
  }
}
Variables
{
  "distinct_on": [
    "column_name"
  ],
  "limit": 123,
  "offset": 123,
  "order_by": [
    "column_name"
  ],
  "where": bool_comparison_exp
}
Response 
{
  "data": {
    "api_trial": [
      {
        "acronym": "abc123",
        "additional_sponsors": jsonb,
        "adverse_events_criteria": "abc123",
        "allocation": "xyz789",
        "arms": [api_arm],
        "basket": true,
        "biomarker_subgroup_analysis": false,
        "end_date": date,
        "endpoints": [api_endpoint],
        "enrollment": 123,
        "indications": jsonb,
        "intervention_model": "abc123",
        "last_update_posted_date": date,
        "masking": "xyz789",
        "num_of_countries": 987,
        "num_of_sites": 987,
        "phase": "xyz789",
        "primary_completion_date": date,
        "primary_sponsors": jsonb,
        "program_trial_arm_cohorts": [
          api_program_trial_arm_cohort
        ],
        "results_first_posted_date": date,
        "results_first_submitted_date": date,
        "safety_acceptable": true,
        "source_id": "abc123",
        "source_link": "xyz789",
        "start_date": date,
        "study_first_submitted_date": date,
        "termination_date": date,
        "title": "xyz789",
        "trial_characteristics": [
          api_trial_characteristics
        ],
        "trial_drugs": [api_trial_drug],
        "trial_id": 987,
        "trial_status": "xyz789",
        "umbrella": true
      }
    ]
  }
}

api_trial_aggregate

Fetch aggregated results for trials.

count, min, max can be used here.

Returns [api_trial_aggregate!]!

Name Description
distinct_on - [String!] distinct select on columns
limit - Int limit the number of rows returned
offset - Int skip the first n rows. Use only with order_by
order_by - [String!] sort the rows by one or more columns
where - bool_comparison_exp filter the rows returned

Example

Query
query api_trial_aggregate($distinct_on: [String!], $limit: Int, $offset: Int, $order_by: [String!], $where: bool_comparison_exp) {
  api_trial_aggregate(distinct_on: $distinct_on, limit: $limit, offset: $offset, order_by: $order_by, where: $where) {
    aggregate {
      ...aggregate_fieldsFragment
    }
    nodes {
      ...api_trialFragment
    }
  }
}
Variables
{
  "distinct_on": [
    "column_name"
  ],
  "limit": 987,
  "offset": 123,
  "order_by": [
    "column_name"
  ],
  "where": bool_comparison_exp
}
Response 
{
  "data": {
    "api_trial_aggregate": [
      {
        "aggregate": aggregate_fields,
        "nodes": [api_trial]
      }
    ]
  }
}

Types

Boolean

The Boolean scalar type represents true or false.

Example
true
Types

Boolean_comparison_exp

Boolean expression to compare columns of type "Boolean". All fields are combined with logical 'AND'.

Input Field Description
_eq - Boolean
_gt - Boolean
_gte - Boolean
_in - [Boolean!]
_is_null - Boolean
_lt - Boolean
_lte - Boolean
_neq - Boolean
_nin - [Boolean!]
Example
{
  "_eq": true,
  "_gt": false,
  "_gte": true,
  "_in": [false],
  "_is_null": true,
  "_lt": false,
  "_lte": false,
  "_neq": false,
  "_nin": [true]
}
Types

Int

The Int scalar type represents non-fractional signed whole numeric values. Int can represent values between -(2^31) and 2^31 - 1.

Example
42
Types

Int_comparison_exp

Boolean expression to compare columns of type "Int". All fields are combined with logical 'AND'.

Input Field Description
_eq - Int
_gt - Int
_gte - Int
_in - [Int!]
_is_null - Boolean
_lt - Int
_lte - Int
_neq - Int
_nin - [Int!]
Example
{
  "_eq": 42,
  "_gt": 42,
  "_gte": 42,
  "_in": [42],
  "_is_null": false,
  "_lt": 42,
  "_lte": 42,
  "_neq": 42,
  "_nin": [42]
}
Types

String

The String scalar type represents textual data, represented as UTF-8 character sequences. The String type is most often used by GraphQL to represent free-form human-readable text.

Types

String_comparison_exp

expression to compare columns of type String. All fields are combined with logical 'AND'.

Input Field Description
_eq - String
_gt - String
_gte - String
_ilike - String
_in - [String!]
_is_null - Boolean
_like - String
_lt - String
_lte - String
_neq - String
_nilike - String
_nin - [String!]
_nlike - String
_nsimilar - String
_similar - String
Example
{
  "_eq": "abc123",
  "_gt": "xyz789",
  "_gte": "xyz789",
  "_ilike": "xyz789",
  "_in": ["xyz789"],
  "_is_null": true,
  "_like": "xyz789",
  "_lt": "xyz789",
  "_lte": "abc123",
  "_neq": "xyz789",
  "_nilike": "abc123",
  "_nin": ["abc123"],
  "_nlike": "abc123",
  "_nsimilar": "xyz789",
  "_similar": "xyz789"
}
Types

aggregate_fields

aggregate fields

Field Name Description
avg - String
count - Int!

Arguments

columns - [String!]

distinct - Boolean

max - String
min - String
Example
{"avg": "abc123", "count": 987, "max": "abc123", "min": "abc123"}
Types

api_arm

A trial arm describes the different groupings of the trial's patients based on the administered drug(s)/intervention(s). Any difference in drug(s)/intervention(s) administered to a patient, be it a different drug, a different combination of drugs, a different dose or even a different treatment schedule constitutes a separate arm.

One arm can be linked to one or multiple programs and can contain one or more cohorts

  • One arm is linked to multiple programs for example when the drug(s)/intervention(s) administered in this arm is tested against multiple indications (Basket trial)
  • An arm may have more than one cohort. For example: Patient cohorts that receive the same drug(s)/intervention(s) will relate to the same arm.
Field Name Description
arm_cohorts - [api_cohort_arm!]! An array relationship
arm_id - Int! Intelligencia's unique arm identifier
arm_name - String Arm name
arm_type - String The arm type describes the role of the drug(s)/intervention(s) that participants receive.
discontinued - Boolean If the trial's arm has been discontinued
program_trial_arm_cohorts - [api_program_trial_arm_cohort!]! An array relationship
therapies - [api_therapy!]! An array relationship
trial_id - Int Intelligencia's unique trial identifier. Each arm belongs to one and only trial_id, but a trial_id may have more than one arm
Example
{
  "arm_cohorts": [api_cohort_arm],
  "arm_id": 123,
  "arm_name": "Trastuzumab",
  "arm_type": "Comparator",
  "discontinued": false,
  "program_trial_arm_cohorts": [
    api_program_trial_arm_cohort
  ],
  "therapies": [api_therapy],
  "trial_id": 123
}
Types

api_benchmarks

Intelligencia's benchmarks consist of the historic approval rate and the number of ongoing and historic (approved, failed) programs (or program-trial pairs) for key clinical development categories. Benchmarks are provided per indication and phase where applicable and are based on data spanning across the last 20 years of clinical development.

Some benchmarks refer to programs, while others refer to trials attached to relevant programs (else called program-trial pairs). For example:

  • Biology benchmarks refer to program biological characteristics (program level benchmarks)
  • Regulatory benchmarks refer to the regulatory designations that a program might have received (program level benchmarks)
  • Trial Design benchmarks refer to the design of clinical trials attached to programs (program-trial pair benchmarks)

Benchmarks have the following dimensions: indication, phase, type, metric, metric value.

Field Name Description
approved - Int Number of approved programs or program-trial pairs
benchmark_category - String Indicates the benchmark's category. For example: Biology, Trial Design or Regulatory
benchmark_id - Int!
benchmark_type - String Indicates if the benchmark counts programs (benchmark_type:program) or program-trial pairs (benchmark_type:trial).
failed - Int Number of failed programs or program-trial pairs
historical - Int Number of historical (approved+failed) programs or program-trial pairs
historical_approval_rate - String % of historical programs or program-trial pairs approved by FDA
indication - String Benchmarks are indication specific and include program/program-trial pairs of descendant indications (see indication hierarchy). Example: benchmarks for pancreatic cancer include all pancreatic cancer subindications, such as pancreatic carcinoma, in the program/program-trial counts.
metric - String Metric is a textual representation of the benchmark metric in question; example: modality, primary endpoints, etc. Empty (NULL) metric values are used for overall benchmarks in the indication and phase dimensions.
metric_value - String The benchmark subcategory. For instance for the benchmark modality, some potential benchmark subcategories are vaccine, antibody etc
ongoing - Int Number of ongoing programs or program-trial pairs
phase - String Indicates the clinical phase that the benchmark refers to. For example assume that the historic approval rate in pancreatic cancer is 18% for the metric=modality and metric_value=small molecule in phase 3. This means that 18% of programs with small molecule drugs that have passed through phase 3 in pancreatic cancer have been approved. For benchmarks across phases, phase is empty (NULL).
Example
{
  "approved": 10,
  "benchmark_category": "Biology",
  "benchmark_id": 987,
  "benchmark_type": "trial",
  "failed": 146,
  "historical": 156,
  "historical_approval_rate": "6%",
  "indication": "Lymphoma",
  "metric": "therapy_type",
  "metric_value": "Combination",
  "ongoing": 187,
  "phase": "Phase 2"
}
Types

api_benchmarks_aggregate

aggregated selection of "api.benchmark"

Field Name Description
aggregate - aggregate_fields
nodes - [api_benchmarks!]!
Example
{
  "aggregate": aggregate_fields,
  "nodes": [api_benchmarks]
}
Types

api_cohort

A cohort separates patients in further subgroupings based on common patient traits e.g. common genetic makeup, age, smoking status etc. Cohorts are used by the investigators for the analysis of the trial's outcomes. A trial's cohort may belong to one or multiple trial arms and it is usually connected to multiple endpoints. For example:

  • One patient cohort is connected with multiple arms when the trial's endpoints are measured for the overall trial population
  • A cohort is usually connected to multiple endpoints per the design of the trial.
Field Name Description
cohort_characteristics - [api_cohort_characteristics!]! An array relationship
cohort_id - Int! Intelligencia's unique cohort identifier
cohort_indication - jsonb The cohort's indications. Along with the cohort's indications we provide information for the evaluation criteria of outcomes relevant to each indication (if provided) and the therapeutic area that the indication belongs to. Cohort's indications are always one or more of the relevant trial's indications.
cohort_name - String Cohort's name
cohort_patients - numeric The number of cohort's participants
cohort_qol - String Cohort's quality of life information
cohort_safety_acceptable - Boolean If the cohort's overall safety profile is acceptable or not
outcomes - [api_outcome!]! An array relationship
program_trial_arm_cohorts - [api_program_trial_arm_cohort!]! An array relationship
publication_source - String If cohort has outcomes, the publication source(s) will be seen here
publish_date - date Date of outcomes seen in cohorts published
trial_id - Int Intelligencia's unique trial identifier
Example
{
  "cohort_characteristics": [api_cohort_characteristics],
  "cohort_id": 123,
  "cohort_indication": [
    {
      "indication": Kaposis sarcoma,
      "therapeutic_area": Oncology,
      "evaluation_criteria": null
    }
  ],
  "cohort_name": "Experimental: 10^6 IV  autologous ova-Tregs",
  "cohort_patients": 155,
  "cohort_qol": "Negative QoL",
  "cohort_safety_acceptable": false,
  "outcomes": [api_outcome],
  "program_trial_arm_cohorts": [
    api_program_trial_arm_cohort
  ],
  "publication_source": "xyz789",
  "publish_date": date,
  "trial_id": 987
}
Types

api_cohort_arm

Provides the relationship between arms and cohorts.

  • A cohort may belong to more than one arm. For example multiple arms are connected with one patient cohort when the trial's endpoints are measured for the overall trial population .
  • An arm may have more than one cohort. For example: Patient cohorts that receive the same drug(s)/intervention(s) will relate to the same arm.
Field Name Description
arm - api_arm! An object relationship
arm_id - Int! Intelligencia's unique arm identifier
cohort - api_cohort! An object relationship
cohort_id - Int! Intelligencia's unique cohort identifier
Example
{
  "arm": api_arm,
  "arm_id": 123,
  "cohort": api_cohort,
  "cohort_id": 987
}
Types

api_cohort_characteristics

Cohort characteristics are all the defining characteristics of a cohort. It includes information that describes the cohorts's patient population, for example line of treatment, stage of disease, molecular characterization, age, sex, smoking status and previous treatments.

Cohort characteristics are organized in criteria groups describing patient groups with homogenous characteristics. For example:

  • group 1: patients with stage 3 disease and EGFR+ gene expression
  • group 2: patients with stage 2 disease and MET+ gene expression

Each unique cohort characteristic is defined by a criterion category (eg. line of treatment), a criterion type (eg. Line I) and a value.

Field Name Description
category - String Criterion categories are: line of treatment, stage of disease, healthy (enrollment of healthy participants), smoking status, sex, indication, previous treated with, patient aging group, molecular, other
cohort - api_cohort An object relationship
comment - String Applicable to other criteria that do not fall under the other criteria categories
criteria_group_id - Int Intelligencia's unique criterion group identifier
criteria_id - Int! Intelligencia's unique criterion identifier
criterion_type - String Criterion types are subcategories that further define each criterion category. For example, criterion types of the criterion category previously treated with are: untreated, drug, other, surgery, radio.
value - String

Value corresponding to the characteristic of the cohort. Expected values depend on the category, and criterion type. For example:

  • Values for criterion category previously treated with and criterion_type drug, indicate the drug name.
  • Values for criterion category molecular and a criterion_type referring to a protein or gene, include options such as mutation, high, medium, low, amplification etc to further characterize the molecular signature of the particular patient criteria group.
  • Values for criterion category Line of Treatment and criterion_type Line II, are Yes/No indicating the inclusion of Line II patients in the cohort.
Example
{
  "category": "xyz789",
  "cohort": api_cohort,
  "comment": "xyz789",
  "criteria_group_id": 123,
  "criteria_id": 987,
  "criterion_type": "xyz789",
  "value": "xyz789"
}
Types

api_drug

Drugs in ongoing clinical programs which are industry-led, FDA-tracked and based on interventional trials, along with all relevant information regarding these drugs, specifically:

  • All names of this drug with preferred name indicated
  • Biology of this drug: modality, mechanism of action (also broken into action type & target), genes, biological_pathways,protein_class
Field Name Description
action_types - jsonb The drug's action type. Action type is part of each drug's mechanism and indicates the effect of the drug on the drug target (eg. agonist, inhibitor etc)
biological_pathways - jsonb The biological pathways that the drug's targets are involved
drug_id - Int! Intelligencia's unique drug identifier
genes - jsonb The drug's target genes. This is relevant for drug targets that are protein-based
mechanism_of_action - jsonb The drug's mechanism(s) of action
modalities - jsonb The drug's modalities. For example: antibody, vaccine, etc.
preferred_name - String The drug's preferred name
program_drugs - [api_program_drug!]! An array relationship
protein_class - jsonb The drug's target(s) protein class(es). Examples: enzyme, receptor, etc.
synonyms - jsonb The drug's synonyms which include drug development names and trade names (if the drug is marketed)
targets - jsonb The drug's targets
treatment_types - jsonb The drug's treatment type. This is an optional field. Example values: chemotherapy, immunotherapy
trial_drugs - [api_trial_drug!]! An array relationship
Example
{
  "action_types": [{"name": Antagonist}],
  "biological_pathways": [
    {"name": Adaptive Immune System},
    {"name": Costimulation by the CD28 family},
    {"name": Immune System},
    {"name": PD-1 signaling}
  ],
  "drug_id": 123,
  "genes": [{"name": PDCD1}],
  "mechanism_of_action": [
    {"name": Programmed cell death protein 1 Antagonist}
  ],
  "modalities": [{"name": Monoclonal antibody}],
  "preferred_name": "Nivolumab",
  "program_drugs": [api_program_drug],
  "protein_class": [{"name": Receptor}],
  "synonyms": [
    {"name": BMS 936558},
    {"name": BMS-936558},
    {"name": BMS-936558-01},
    {"name": BMS-986298},
    {"name": MDX 1106},
    {"name": MDX-1106},
    {"name": NIVO},
    {"name": Nivolumab},
    {"name": Nivolumab BMS},
    {"name": ONO-0123},
    {"name": ONO 4538},
    {"name": ONO-4538},
    {"name": Opdivo}
  ],
  "targets": [{"name": Programmed cell death protein 1}],
  "treatment_types": [{"name": Immunotherapy}],
  "trial_drugs": [api_trial_drug]
}
Types

api_drug_aggregate

aggregated selection of "api.drug"

Field Name Description
aggregate - aggregate_fields
nodes - [api_drug!]!
Example
{
  "aggregate": aggregate_fields,
  "nodes": [api_drug]
}
Types

api_endpoint

The endpoints of a trial. An endpoint is a metric of the outcomes of the trial. Endpoints are usually part of the clinical trial design. When a trial starts the information that is usually available includes the number and type/subtype of endpoints. Later in the trial, when outcomes are published e.g. in papers, conferences, company announcements, then clinical endpoints are connected with the relevant results (outcomes). An endpoint is defined by a type and a subtype (when applicable). For example: for endpoint type "safety" there are several subtypes showing the type of adverse event assessment such as "Grade 2 AEs", "Grade 3 AEs" etc There are cases where the endpoint subtype is not provided such as:

  • When this endpoint has not applicable subtypes
  • When this endpoint has subtypes but it can also be measured as a standalone endpoint type e.g. PASI.
Field Name Description
comment - String Name of Endpoint where Endpoint type: Other
endpoint_id - Int! Intelligencia's unique endpoint identifier
endpoint_subtype - String The endpoint subcategorization (when applicable)
endpoint_type - String The endpoint name. Example: ORR, OS, PFS etc
endpoint_value_unit - String The endpoint's value unit
primary_choice - String If the endpoint is primary, secondary or other in this cohort
timeframe - String The endpoint's monitoring time period.
timeframe_name - String The endpoints timeframe unit (example: months, weeks, etc). When the value is overall, it means that the endpoint is monitored throughout the trial's duration.
trial - api_trial An object relationship
trial_id - Int Intelligencia's unique trial identifier
Example
{
  "comment": "xyz789",
  "endpoint_id": 123,
  "endpoint_subtype": "Period",
  "endpoint_type": "PFS",
  "endpoint_value_unit": "Percentage",
  "primary_choice": "Primary",
  "timeframe": "12",
  "timeframe_name": "months",
  "trial": api_trial,
  "trial_id": 123
}
Types

api_indication_hierarchy

Describes Intelligencia's expertly curated indication ontology for the therapeutic areas that the supported indications belong to (Oncology, Immunology/Inflammation and CNS). The indication ontology is a multilevel hierarchy tree based on parent-child relationships. Ancestor-descendant indication relationships must be derived. For example:

  • The indication Triple negative breast neoplasms is a child of breast carcinoma, breast carcinoma is a child of breast cancer, therefore Triple negative breast neoplasms is a descendant of breast cancer.
Field Name Description
child_level - Int Child's distance from the indication hierarchy root. Example: indication_child: Multiple sclerosis child_level: 2
distance - Int Distance between parent and child in the indication tree. Example: 2
indication_child - String Modality first-level child name. Example: Bispecific antibody
indication_parent - String Indication parent name. Example: Breast Cancer
parent_level - Int Parent's distance from the indication hierarchy root. Example: indication_parent: Neurodegenerative diseases parent_level: 1
Example
{
  "child_level": 987,
  "distance": 123,
  "indication_child": "Breast carcinoma",
  "indication_parent": "Breast cancer",
  "parent_level": 123
}
Types

api_modality_hierarchy

Intelligencia's expertly curated drug modality ontology. The modality ontology is a multilevel hierarchy tree based on parent-child relationships. Example: The modality CAR T cells is a child modality of Cell Therapy.

Field Name Description
child_level - Int Child's distance from the modality hierarchy root. Example: indication_child: Nonviral-vector gene therapy child_level: 2
distance - Int Distance between parent and child in the indication tree. Example: 2
modality_child - String Modality first-level child name. Example: Bispecific antibody
modality_parent - String Modality parent name. Example: Antibody
parent_level - Int Parent's distance from the modality hierarchy root. Example: indication_parent: Gene therapy parent_level: 1
Example
{
  "child_level": 123,
  "distance": 123,
  "modality_child": "Trastuzumab",
  "modality_parent": "abc123",
  "parent_level": 123
}
Types

api_outcome

The trial outcomes of a specific cohort. A cohort typically contains outcome measures for more than one endpoint (either primary, secondary or other). Cohort outcomes are filled with the results of the corresponding endpoints when trial outcomes are published e.g. in papers, conferences, company announcements.

Field Name Description
cohort - api_cohort An object relationship
cohort_id - Int! Intelligencia's unique cohort identifier
confidence_interval_high - numeric Highest confidence interval of specific endpoint
confidence_interval_low - numeric Lowest confidence interval of specific endpoint
confidence_interval_percent - numeric Confidence level at which endpoint was assessed
endpoint - api_endpoint An object relationship
endpoint_id - Int! Intelligencia's unique endpoint identifier
endpoint_scope - String Indicates whether the endpoint was part of the initial trial's design, or resulted as part of the trial's outcomes analysis stage
endpoint_value - String The endpoint outcome value
independent - Boolean FALSE: Endpoint assessed by Investigator TRUE: Endpoint assessed by Independent Review Committee
outcome_id - Int! Intelligencia's unique outcome identifier
Example
{
  "cohort": api_cohort,
  "cohort_id": 123,
  "confidence_interval_high": numeric,
  "confidence_interval_low": numeric,
  "confidence_interval_percent": numeric,
  "endpoint": api_endpoint,
  "endpoint_id": 123,
  "endpoint_scope": "abc123",
  "endpoint_value": "abc123",
  "independent": true,
  "outcome_id": 987
}
Types

api_outcome_comparison

The trial outcome comparison analysis of a specific cohort. Outcome value comparison is relevant only to some endpoints. Possible values are "baseline", "placebo", "comparator". Specifically: When value is "Baseline": difference between the endpoint measurement at the time of capturing the endpoint vs the measurement of the same endpoint at an earlier time, usually the initial measurement before treatment start. When value is either "placebo" or "comparator": the difference between the endpoint value at the given timepoint for the experimental cohort from the endpoint for the placebo or comparator cohort is provided.

Field Name Description
cohort - api_cohort An object relationship
cohort_id - Int Intelligencia's unique cohort identifier
comparator_cohort - api_cohort An object relationship
comparator_cohort_id - Int Cohort that comparison is done with
comparison_ci_high - numeric Highest confidence interval of endpoint comparison
comparison_ci_low - numeric Lowest confidence interval of endpoint comparison
comparison_ci_percent - numeric Confidence level at which endpoint comparison is assessed
comparison_description - String Possible values are baseline, placebo, comparator
comparison_p_value - numeric p-value for comparison of specific endpoint
comparison_p_value_operator - String Operator of p-value (Equals, Less than, More than)
comparison_type - String Comparison type for the specific endpoint; ratio, difference, etc
comparison_value - numeric Comparison value for the specific endpoint
endpoint - api_endpoint An object relationship
endpoint_id - Int Intelligencia's unique endpoint identifier
outcome_comparison_id - Int! Intelligencia's unique outcome comparison identifier
Example
{
  "cohort": api_cohort,
  "cohort_id": 123,
  "comparator_cohort": api_cohort,
  "comparator_cohort_id": 987,
  "comparison_ci_high": numeric,
  "comparison_ci_low": numeric,
  "comparison_ci_percent": numeric,
  "comparison_description": "abc123",
  "comparison_p_value": numeric,
  "comparison_p_value_operator": "xyz789",
  "comparison_type": "xyz789",
  "comparison_value": numeric,
  "endpoint": api_endpoint,
  "endpoint_id": 123,
  "outcome_comparison_id": 123
}
Types

api_phase_hierarchy

Describes the hierarchy of clinical trial phases which is based on parent-child relationships. Example: Phase 1b is a child of phase 1

Field Name Description
phase_child - String Clinical phase first-level child name. Example: Phase 1b
phase_parent - String Clinical phase parent name. Example: Phase 1
Example
{
  "phase_child": "Phase 1b",
  "phase_parent": "Phase 1"
}
Types

api_program

A program (also known as clinical pipeline or drug pipeline) is the clinical development of a drug (or a set of drugs in case of combination therapies) by a pharmaceutical company (alone or in collaboration with other partners) for a specific indication and specific patient profile. A program consists of a set of clinical trials with the ultimate goal of approval for marketing.

Programs are defined based on their drug(s)/interventions, indication, primary sponsor, adjuvant value and program patient characteristics. A program's latest phase is derived from the program's latest most advanced clinical trial phase.

Field Name Description
adjuvant - jsonb Indicates if the program's therapy is either adjuvant and/or neoadjuvant when applicable
administration_mode - citext The program's primary drug route of administration in the human body. Example: Oral, Injectable
end_date - date This is the end date of the program's trial at the most advanced clinical phase with the latest completion date.
indication - citext The program's indication (example: breast carcinoma)
intervention - jsonb Indicates if the program includes other non-drug related interventions such as radiotherapy, surgery, photodynamic therapy etc
latest_trial - Int The unique clinical trial identifier of the program's latest clinical phase trial. This identifier can be used in conjunction with the clinical trial entity to find further information about this Program's latest trial. Latest is defined as the trial that has the most advanced clinical phase and the latest end date.
latest_trial_with_outcomes - Int The unique clinical trial identifier of the program's latest clinical trial with published outcomes. This identifier can be used in conjunction with the clinical trial entity to find further information about this Program's latest trial with outcomes. Latest with outcomes is defined as the trial that has the most advanced clinical phase and latest end date with outcomes published.
link_to_platform - String The program's link to Intelligencia's Insight platform and specifically at the program's deep dive section
primary_sponsor - citext The program's primary sponsor company name (example: Pfizer)
program_characteristics - [api_program_characteristics!]! An array relationship
program_drivers - [api_program_drivers!]! An array relationship
program_drugs - [api_program_drug!]! An array relationship
program_id - Int! Intelligencia's unique program identifier
program_latest_trial - api_trial An object relationship
program_latest_trial_with_outcomes - api_trial An object relationship
program_ptrs - api_program_ptrs An object relationship
program_trial_arm_cohorts - [api_program_trial_arm_cohort!]! An array relationship
program_trial_pems - [api_program_trial_pem!]! An array relationship
regulatory_pathways - jsonb Indicates if the program has received a facilitated regulatory designation from FDA such as orphan drug, breakthrough therapy, fast track, priority review
start_date - date Start date of program. In cases of multiple trials, program start date is the earliest trial start date
status - String
therapeutic_area - jsonb The generalized group of diseases that the program's indication belongs to. Note that some indications might belong to more than 1 TAs. However, every program belongs to a single indication and a single TA.
therapy_type - citext Indicates if the program's therapy is monotherapy or combination. In the case of combination therapies one drug is the primary drug per program and the rest are the additional drugs.
Example
{
  "adjuvant": [{"name": Neoadjuvant}],
  "administration_mode": Injectable/Intravenous,
  "end_date": 2023-12-08,
  "indication": Leukemia, myeloid, acute,
  "intervention": [{"name": Radio}],
  "latest_trial": 789,
  "latest_trial_with_outcomes": 789,
  "link_to_platform": "https://insight.intelligencia.ai/insight/portfolio-optimizer/program-assessment/deep-dive#programId=986",
  "primary_sponsor": Jasper Therapeutics,
  "program_characteristics": [
    api_program_characteristics
  ],
  "program_drivers": [api_program_drivers],
  "program_drugs": [api_program_drug],
  "program_id": 986,
  "program_latest_trial": api_trial,
  "program_latest_trial_with_outcomes": api_trial,
  "program_ptrs": api_program_ptrs,
  "program_trial_arm_cohorts": [
    api_program_trial_arm_cohort
  ],
  "program_trial_pems": [api_program_trial_pem],
  "regulatory_pathways": [{"name": Orphan drug}],
  "start_date": date,
  "status": "Ongoing",
  "therapeutic_area": [{"name": Oncology}],
  "therapy_type": Combination
}
Types

api_program_aggregate

aggregated selection of "api.program"

Field Name Description
aggregate - aggregate_fields
nodes - [api_program!]!
Example
{
  "aggregate": aggregate_fields,
  "nodes": [api_program]
}
Types

api_program_benchmark

columns and relationships of "api.program_benchmark"

Field Name Description
benchmark - api_benchmarks An object relationship
benchmark_id - Int
program - api_program An object relationship
program_id - Int
Example
{
  "benchmark": api_benchmarks,
  "benchmark_id": 987,
  "program": api_program,
  "program_id": 987
}
Types

api_program_characteristics

Program characteristics are all the defining characteristics of a clinical program beyond its drug(s)/intervention(s), indication, primary sponsor and adjuvant value. It includes information that describes the program's patient population, for example line of treatment, stage of disease, molecular characterization, age, sex, smoking status and previous treatments.

Program characteristics are organized in criteria groups describing patient groups with homogenous characteristics. For example:

  • group 1: patients with stage 3 disease and EGFR+ gene expression
  • group 2: patients with stage 2 disease and MET+ gene expression

Each unique program characteristic is defined by a criterion category (eg. line of treatment), a criterion type (eg. Line I) and a value.

Field Name Description
category - String Criterion categories are: line of treatment, stage of disease, healthy (enrollment of healthy participants), smoking status, sex, indication, previous treated with, patient aging group, molecular, other
comment - String Applicable to other criteria that do not fall under the other criteria categories
criteria_group_id - Int Intelligencia's unique criterion group identifier
criteria_id - Int! Intelligencia's unique criterion identifier
criterion_type - String Criterion types are subcategories that further define each criterion category. For example, criterion types of the criterion category previously treated with are: untreated, drug, other, surgery, radio.
program - api_program An object relationship
value - String

Value corresponding to the characteristic of the program. Expected values depend on the category, and criterion type. For example:

  • Values for criterion category previously treated with and criterion_type drug, indicate the drug name.
  • Values for criterion category molecular and a criterion_type referring to a protein or gene, include options such as mutation, high, medium, low, amplification etc to further characterize the molecular signature of the particular patient criteria group.
  • Values for criterion category Line of Treatment and criterion_type Line II, are Yes/No indicating the inclusion of Line II patients in the program's clinical trials.
Example
{
  "category": "Molecular",
  "comment": "xyz789",
  "criteria_group_id": 987,
  "criteria_id": 123,
  "criterion_type": "Receptor tyrosine-protein kinase erbB-2",
  "program": api_program,
  "value": "Positive"
}
Types

api_program_drivers

Program's PTRS drivers and relevant historic approval rates. Drivers are the features that the ML algorithms take into account to predict the program's PTRS. Intelligencia's PTRS drivers belong to feature families such as Biology, Clinical Trial Outcomes, Clinical Trial Design, Regulatory, Company Characteristics. Each PTRS driver has a boolean or numeric type (see feature_type field) and a value for the related program. For instance,

  • Randomized trial indicates if the trial uses randomization, and the program values are 0/1 (boolean feature).
  • number of sites indicates the number of trial sites, and the program values are numeric (numeric feature). Certain drivers refer to program level data (otherwise called program level drivers), and other drivers refer to trial specific data (otherwise called program-trial pair drivers). Program-trial pair drivers appear multiple times if the program selected has more than one trial. Specifically:
  • Biology drivers relate to the biological characteristics of the program (program level drivers)
  • Clinical trial outcomes relate to the outcomes of every trial attached to this program (program-trial pair drivers)
  • Clinical Trial Design drivers relate to the design of every trial attached to this program (program-trial pair drivers)
  • Regulatory drivers relate to the regulatory designations that a program might have received (program level drivers)
  • Company characteristics relate to the company sponsor of each trials attached to this program (program-trial pair drivers)

PTRS driver historic approval rate is the % of historic programs/program-trial pairs having similar program values in this indication approved by FDA. Specifically:

  • For program level drivers: the historic approval rate refers to programs in the same indication with values similar to the selected program. For example: for the PTRS driver drug's action type: antagonist and program value 1 (TRUE) and indication breast cancer, historic approval rate is calculated as the approved breast cancer programs with action type antagonist over all historic breast cancer programs with the same action type.
  • For program-trial pair drivers: the historic approval rate refers to program-trial pairs with values similar to the relevant program-trial pairs at this indication and phase. For example: for the PTRS driver Drug administration: Oral and program-trial value 1 (TRUE) and indication breast cancer and phase 2, the historic approval rate is calculated as the approved breast cancer phase 2 program-trial pairs with oral drug administration over all historic breast cancer program-trial pairs with oral drug administration at phase Similar program or trial values are defined as follows:
  • For features with boolean type, these are the exact programs (or program-trial pairs) values
  • For features with numeric values, similar program (or program-trial pair) values are identified by a classification algorithm that determines ranges of values per feature that exhibit similar historical approval behavior. In cases where there is a limited number of programs (or program-trial pairs) with similar values, the approval rate is not available.
Field Name Description
feature_type - String If the feature is boolean or numeric
historical_approval_rate - numeric % of historical programs/program-trial pairs having similar program values in this indication approved by FDA. In cases where there is a limited number of programs (or trials) with similar value, the approval rate is not available
indication - String The indication for which each PTRS driver's historic approval rate refers to
program_id - Int Intelligencia's unique program identifier.
program_value - numeric The program's value for the PTRS driver
ptrs_feature_description - String PTRS driver name
ptrs_feature_family - String Intelligencia's PTRS drivers are categorized in feature families: Biology, Clinical Trial Design, Regulatory, Clinical Trial Outcomes, Company Characteristics
trial_id - Int Intelligencia's unique trial identifier for the program's trial that the specific driver refers to. This is relevant only for program-trial pair drivers: Clinical Trial design, Clinical Trial Outcomes and Company Characteristics
Example
{
  "feature_type": "boolean",
  "historical_approval_rate": numeric,
  "indication": "Stomach cancer",
  "program_id": 123,
  "program_value": 1,
  "ptrs_feature_description": "Drug's action type group: Blocker",
  "ptrs_feature_family": "Biology",
  "trial_id": 987
}
Types

api_program_drug

Links a program with its related drugs. Specifically this entity links the program's program_id with the program's primary drug ID and additional drug IDs (in case of combination therapies). A program's primary drug is defined as the program's main experimental drug.

Field Name Description
drug - api_drug! An object relationship
drug_id - Int! Intelligencia's unique drug identifier
is_primary - Boolean true=the drug is the program's primary drug, false=the drug is part of the combination therapy of this program but it is not the primary drug
program - api_program! An object relationship
program_id - Int! Intelligencia's unique program identifier
Example
{
  "drug": api_drug,
  "drug_id": 123,
  "is_primary": false,
  "program": api_program,
  "program_id": 123
}
Types

api_program_ptrs

Contains Intelligencia's Probability of Phase Transition and Probability of Technical and Regulatory Success (PTRS) of ongoing, interventional, industry-sponsored, clinical development programs.

  • Probability of phase transition is the probability of a program to move to the next phase.

  • There are cases where PTRS or Probability of phase transition are not provided due to insufficient data. In those cases, the program will not appear in this table Program’s PTRS is compared to other similar ongoing programs’ PTRS and categorized in a quarter i.e.PTRS of the fourth quarter means that the program PTRS is better than the PTRS of the 75% of ongoing programs with similar characteristics.

  • Program similarity is defined based on the indication family and development stage. If there is not a sufficient number of programs, there will be no PTRS quarter or quartiles showing up but similar programs will still be available

Field Name Description
development_stage - String

Program's latest development stage for the purpose of comparing it with other similar programs

Value: Phase X started i.e. Phase 1 started Description: These programs are at the start of the latest phase trial.

Value: Phase X; public outcomes not available Description: Public outcomes for the major endpoints are unavailable for the latest phase trial of these programs.

Value: Phase X with public outcomes Description: The latest phase trial of these programs has available outcomes for the major endpoints
phase_transition_probability - String Program's probability to transition to the next clinical phase.
program - api_program An object relationship
program_id - Int! Intelligencia's unique program identifier
program_related_programs - [api_program_related_programs!]! An array relationship
ptrs - String Program's Probability of Technical and Regulatory Success
ptrs_breakdown - [api_program_ptrs_breakdown!]! An array relationship
ptrs_indication - String
ptrs_quarter - Int Program's PTRS quarter compared to other similar ongoing programs. Values: 1, 2, 3, 4
ptrs_quartiles - jsonb Quartiles are calculated based on the PTRS of all similar programs. The value describes the upper end of the quartile. Example. {"1": 20, "2": 35, "3": 43}. 20 is the PTRS value under which 25% of programs are found when arranged in increasing order.
Example
{
  "development_stage": "Phase 2; public outcomes not available",
  "phase_transition_probability": "79%",
  "program": api_program,
  "program_id": 123,
  "program_related_programs": [
    api_program_related_programs
  ],
  "ptrs": "20%",
  "ptrs_breakdown": [api_program_ptrs_breakdown],
  "ptrs_indication": "Non-Small cell lung carcinoma",
  "ptrs_quarter": 1,
  "ptrs_quartiles": {"1": 2, "2": 6, "3": 10}
}
Types

api_program_ptrs_breakdown

Contains phase-level transition probabilities for ongoing clinical programs, derived from Intelligencia AI’s PTRS modeling framework. Each value represents the probability of advancing from a specific development phase to its subsequent one in the clinical development pathway, based on program-level characteristics.

Field Name Description
phase - String! Clinical development phase at which the transition probability is assigned (e.g., "Phase 1", "Phase 3", "Registration")
phase_level_transition_probability - String AI-driven probability for the program to progress from the specified phase to the subsequent one in the development pathway
program_id - Int! Intelligencia's unique program identifier
Example
{
  "phase": "Phase 1",
  "phase_level_transition_probability": "20%",
  "program_id": 789
}
Types

api_program_trial_arm_cohort

Links a program with its trials, arms and cohorts.

  • A program may have more than one trial.
  • A trial may be linked to more than one program. A trial may be linked to more than one program for example when a) a trial tests the same drug(s)/intervention(s) in multiple indications (Basket Trial), b) a trial tests different drug(s)/intervention(s) in the same indication (Umbrella Trial), c) a trial that tests multiple subsets of the same indication (e.g subsets of breast cancer), or multiple patient groups under the same indication. Note that a trial can be both basket and umbrella, namely it can test multiple different drug therapies in multiple indications.
  • A trial may have more than one arm. Specifically, any difference in drug(s)/intervention(s) administered to a patient, be it a different drug, a different combination of drugs, a different dose or even a different treatment schedule constitutes a separate arm.
  • An arm may have more than one cohort. Specifically, a cohort is a patient subgrouping which is based on any common trait that patients may have, for instance common genetic makeup, age, smoking status etc. So patient subgroups that receive the same drug(s)/intervention(s) will relate to the same arm.
  • A cohort may be linked to more than one arm. For example, multiple arms are connected with one patient cohort when the trial's endpoints are measured for the overall trial population.
Field Name Description
arm - api_arm An object relationship
arm_id - Int Intelligencia's unique arm identifier
cohort - api_cohort An object relationship
cohort_id - Int Intelligencia's unique cohort identifier
program - api_program An object relationship
program_id - Int Intelligencia's unique program identifier
trial - api_trial An object relationship
trial_id - Int Intelligencia's unique trial identifier
Example
{
  "arm": api_arm,
  "arm_id": 123,
  "cohort": api_cohort,
  "cohort_id": 123,
  "program": api_program,
  "program_id": 123,
  "trial": api_trial,
  "trial_id": 987
}
Types

api_program_trial_pem

Captures information on whether a trial met its primary endpoints for a specific phase.

Field Name Description
is_pem - Boolean Trial's primary endpoint(s) were met/not met or information could not be inferred for that program
phase - String! Phase of trial that primary endpoint met information corresponds to
primary_endpoint_met_type - String Trial has a singular, multiple or co-primary endpoints
program - api_program! An object relationship
program_id - Int! Intelligencia's unique program identifier
source - String Source of endpoint met information if that was concluded from a paper/announcement
source_type - String Endpoint met information was inferred or concluded from a paper/announcement
trial - api_trial! An object relationship
trial_id - Int! Intelligencia's unique trial identifier
Example
{
  "is_pem": false,
  "phase": "xyz789",
  "primary_endpoint_met_type": "xyz789",
  "program": api_program,
  "program_id": 987,
  "source": "xyz789",
  "source_type": "xyz789",
  "trial": api_trial,
  "trial_id": 987
}
Types

api_target

Contains targets along with information about their synonyms, related organism and target type (eg. protein, gene). A drug target is a molecule in the body, usually a protein, that is intrinsically associated with a particular disease process and that could be addressed by a drug to produce a desired therapeutic effect. Drugs can have multiple targets.

Field Name Description
name - citext Intelligencia's unique program identifier
organism - citext Target's organism i.e. Human, House Mouse
synonyms - jsonb Target's synonyms
target_id - Int! Intelligencia's unique program identifier
type - citext Target's type i.e. Protein, Gene
Example
{
  "name": Homeobox protein DLX-5,
  "organism": human,
  "synonyms": [{"name": Homeobox protein DLX-5}],
  "target_id": 234,
  "type": Protein
}
Types

api_therapy

A therapy is a selection of drugs and interventions administered to a particular patient group (arm participants) on scheduled intervals for the duration of a set period of time. In cases of multi-period arms, order specifies the sequence of therapies' administration to patients.

Field Name Description
additional_drugs - jsonb Additional drugs included in the therapy
arm - api_arm An object relationship
arm_id - Int! Intelligencia's unique arm identifier
discontinued - Boolean Indicates whether a therapy was discontinued before the planned end date
drug_schedule - String The schedule interval of the therapy's administration
interventions - jsonb Additional interventions besides the drug(s) listed
order - Int The sequence order that this therapy was administered to patients
patient_number - Int
primary_drug - String Therapy's primary drug
primary_drug_administration_mode - String Administration mode with which drug was given to participants
primary_drug_dosage - numeric Dosage of given primary drug
primary_drug_dosage_unit - String Dosage unit of given primary drug
primary_drug_titration_ending_dose - numeric The primary drug's final titration dose (if applicable)
primary_drug_titration_starting_dose - numeric The primary drug's starting titration dose (if applicable)
therapy_id - Int! Intelligencia's unique therapy identifier
therapy_type - String Describes the role of the drug(s)/intervention(s) that participants receive
Example
{
  "additional_drugs": jsonb,
  "arm": api_arm,
  "arm_id": 123,
  "discontinued": true,
  "drug_schedule": "xyz789",
  "interventions": jsonb,
  "order": 987,
  "patient_number": 123,
  "primary_drug": "xyz789",
  "primary_drug_administration_mode": "abc123",
  "primary_drug_dosage": numeric,
  "primary_drug_dosage_unit": "abc123",
  "primary_drug_titration_ending_dose": numeric,
  "primary_drug_titration_starting_dose": numeric,
  "therapy_id": 987,
  "therapy_type": "xyz789"
}
Types

api_trial

Clinical trials that relate to currently ongoing programs. A clinical trial is a research study conducted in humans which assesses a treatment or a medical intervention. Intelligencia's database includes interventional, industry-led clinical trials which are FDA-tracked and part of ongoing clinical programs.

A trial may be related to more than one program and each trial may have more than one arm. Specifically,

  • A trial may be linked to more than one program for example when a trial tests the same drug(s)/intervention(s) in multiple indications (Basket Trial)
  • During a trial, any difference in drug(s)/intervention(s) administered to a patient - be it a different drug, a different combination of drugs, a different dose or even a different treatment schedule - constitutes a separate arm.
Field Name Description
acronym - String Popular trial name corresponding to trial
additional_sponsors - jsonb The trial's additional sponsors
adverse_events_criteria - String The type of adverse events criteria the trial uses to report adverse events
allocation - String Indicates if the trial is randomized or not
arms - [api_arm!]! An array relationship
basket - Boolean A basket trial is a clinical trial that tests a specific drug or drug combination/intervention(s) in multiple separate indications. True= the trial is basket, False= the trial is not basket
biomarker_subgroup_analysis - Boolean True when any of the trial's arms' contains cohorts that are defined based on a biomarker (most common cases of biomarkers can be molecular, previous treatments or age)
end_date - date The trial's completion date as in clinicaltrials.gov. If the trial's completion date is missing then the field has the trial's primary completion date as in clinicaltrials.gov
endpoints - [api_endpoint!]! An array relationship
enrollment - Int The trial's participant population
indications - jsonb The trial's indications. Along with the trial's indications we provide information for the evaluation criteria of outcomes relevant to each indication (if provided) and the therapeutic area that the indication belongs to
intervention_model - String
last_update_posted_date - date Date that the latest update regarding a trial was posted on clinicaltrials.gov
masking - String The type of masking that is implemented in the trial
num_of_countries - Int The trial's number of site countries (as seen in clinicaltrials.gov)
num_of_sites - Int The trial's number of sites (as seen in clinicaltrials.gov)
phase - String The trial's clinical phase
primary_completion_date - date Date that the final subject was examined or received an intervention for purposes of the final collection of data for the primary outcome measure
primary_sponsors - jsonb The trial's primary sponsors
program_trial_arm_cohorts - [api_program_trial_arm_cohort!]! An array relationship
results_first_posted_date - date Date that results of trial where first posted in clinicaltrials.gov (Note: results shown in endpoint values might differ from ones seen in clinicaltrials.gov, if more updated paper has been published)
results_first_submitted_date - date Date that results of trial where first submitted in clinicaltrials.gov (Note: results shown in endpoint values might differ from ones seen in clinicaltrials.gov, if more updated paper has been published)
safety_acceptable - Boolean If the trial's overall safety profile is acceptable or not
source_id - String Unique trial's identifier as in clinicaltrials.gov. For trials that are not registered in clinicaltrials.gov because they are very old, the source_id starts with the IAI- prefix
source_link - String The link to clinicaltrials.gov for each trial
start_date - date The trial's start date
study_first_submitted_date - date Date that trial was first submitted to clinicaltrials.gov
termination_date - date The termination date of trials with status terminated
title - String The trial's title as it appears in clinicaltrials.gov. For trials with the IAI- prefix this is absent
trial_characteristics - [api_trial_characteristics!]! An array relationship
trial_drugs - [api_trial_drug!]! An array relationship
trial_id - Int! Intelligencia's unique trial identifier
trial_status - String
umbrella - Boolean An umbrella trial is a clinical trial that tests multiple different drug(s)/intervention(s) in one specific indication. True= the trial is umbrella, False= the trial is not umbrella
Example
{
  "acronym": "abc123",
  "additional_sponsors": jsonb,
  "adverse_events_criteria": "xyz789",
  "allocation": "abc123",
  "arms": [api_arm],
  "basket": false,
  "biomarker_subgroup_analysis": true,
  "end_date": date,
  "endpoints": [api_endpoint],
  "enrollment": 987,
  "indications": jsonb,
  "intervention_model": "xyz789",
  "last_update_posted_date": date,
  "masking": "xyz789",
  "num_of_countries": 123,
  "num_of_sites": 123,
  "phase": "xyz789",
  "primary_completion_date": date,
  "primary_sponsors": jsonb,
  "program_trial_arm_cohorts": [
    api_program_trial_arm_cohort
  ],
  "results_first_posted_date": date,
  "results_first_submitted_date": date,
  "safety_acceptable": true,
  "source_id": "abc123",
  "source_link": "xyz789",
  "start_date": date,
  "study_first_submitted_date": date,
  "termination_date": date,
  "title": "abc123",
  "trial_characteristics": [api_trial_characteristics],
  "trial_drugs": [api_trial_drug],
  "trial_id": 123,
  "trial_status": "xyz789",
  "umbrella": false
}
Types

api_trial_aggregate

aggregated selection of "api.trial"

Field Name Description
aggregate - aggregate_fields
nodes - [api_trial!]!
Example
{
  "aggregate": aggregate_fields,
  "nodes": [api_trial]
}
Types

api_trial_characteristics

Trial characteristics are all the defining characteristics of a clinical trial beyond its drug(s)/intervention(s), indication, primary sponsor and adjuvant value. It includes information that describes the trial's patient population, for example line of treatment, stage of disease, molecular characterization, age, sex, smoking status and previous treatments.

Trial characteristics are organized in criteria groups describing patient groups with homogenous characteristics. For example:

  • group 1: patients with stage 3 disease and EGFR+ gene expression
  • group 2: patients with stage 2 disease and MET+ gene expression

Each unique trial characteristic is defined by a criterion category (eg. line of treatment), a criterion type (eg. Line I) and a value.

Field Name Description
category - String Criterion categories are: line of treatment, stage of disease, healthy (enrollment of healthy participants), smoking status, sex, indication, previous treated with, patient aging group, molecular, other
comment - String Applicable to other criteria that do not fall under the other criteria categories
criteria_group_id - Int Intelligencia's unique criterion group identifier
criteria_id - Int! Intelligencia's unique criterion identifier
criterion_type - String Criterion types are subcategories that further define each criterion category. For example, criterion types of the criterion category previously treated with are: untreated, drug, other, surgery, radio.
trial - api_trial An object relationship
trial_id - Int Intelligencia's unique trial identifier
value - String

Value corresponding to the characteristic of the trial. Expected values depend on the category, and criterion type. For example:

  • Values for criterion category previously treated with and criterion_type drug, indicate the drug name.
  • Values for criterion category molecular and a criterion_type referring to a protein or gene, include options such as mutation, high, medium, low, amplification etc to further characterize the molecular signature of the particular patient criteria group.
  • Values for criterion category Line of Treatment and criterion_type Line II, are Yes/No indicating the inclusion of Line II patients in the clinical trial.
Example
{
  "category": "abc123",
  "comment": "abc123",
  "criteria_group_id": 123,
  "criteria_id": 123,
  "criterion_type": "abc123",
  "trial": api_trial,
  "trial_id": 123,
  "value": "abc123"
}
Types

api_trial_drug

Provides the relationship between trials and drugs. One trial can have one drug (monotherapy) or multiple drugs administered (combination therapy)

Field Name Description
drug - api_drug! An object relationship
drug_id - Int! Intelligencia's unique drug identifier
is_primary - Boolean! True: when the drug is one of the main experimental drugs in the trial, False: the drug is not one of the main experimental drug of the trial
trial - api_trial! An object relationship
trial_id - Int! Intelligencia's unique trial identifier
Example
{
  "drug": api_drug,
  "drug_id": 123,
  "is_primary": false,
  "trial": api_trial,
  "trial_id": 987
}
Types

bool_comparison_exp

Boolean expression to filter rows from any api type . All fields are combined with a logical 'AND'.

Input Field Description
_and - [bool_comparison_exp!]
_not - bool_comparison_exp
_or - [bool_comparison_exp!]
boolean_field - Boolean_comparison_exp
jsonb_column - jsonb_comparison_exp
citext_column - citext_comparison_exp
int_column - Int_comparison_exp
numeric_column - numeric_comparison_exp
relation - bool_comparison_exp
Example
{
  "_and": [bool_comparison_exp],
  "_not": bool_comparison_exp,
  "_or": [bool_comparison_exp],
  "boolean_field": Boolean_comparison_exp,
  "jsonb_column": jsonb_comparison_exp,
  "citext_column": citext_comparison_exp,
  "int_column": Int_comparison_exp,
  "numeric_column": numeric_comparison_exp,
  "relation": bool_comparison_exp
}
Types

citext

Example
object
Types

citext_comparison_exp

Boolean expression to compare columns of type "citext". All fields are combined with logical 'AND'.

Input Field Description
_eq - citext
_gt - citext
_gte - citext
_ilike - citext

does the column match the given case-insensitive pattern
_in - [citext!]
_iregex - citext

does the column match the given POSIX regular expression, case insensitive
_is_null - Boolean
_like - citext

does the column match the given pattern
_lt - citext
_lte - citext
_neq - citext
_nilike - citext

does the column NOT match the given case-insensitive pattern
_nin - [citext!]
_niregex - citext

does the column NOT match the given POSIX regular expression, case insensitive
_nlike - citext

does the column NOT match the given pattern
_nregex - citext

does the column NOT match the given POSIX regular expression, case sensitive
_nsimilar - citext

does the column NOT match the given SQL regular expression
_regex - citext

does the column match the given POSIX regular expression, case sensitive
_similar - citext

does the column match the given SQL regular expression
Example
{
  "_eq": citext,
  "_gt": citext,
  "_gte": citext,
  "_ilike": citext,
  "_in": [citext],
  "_iregex": citext,
  "_is_null": true,
  "_like": citext,
  "_lt": citext,
  "_lte": citext,
  "_neq": citext,
  "_nilike": citext,
  "_nin": [citext],
  "_niregex": citext,
  "_nlike": citext,
  "_nregex": citext,
  "_nsimilar": citext,
  "_regex": citext,
  "_similar": citext
}
Types

date

Example
object
Types

date_comparison_exp

Boolean expression to compare columns of type "date". All fields are combined with logical 'AND'.

Input Field Description
_eq - date
_gt - date
_gte - date
_in - [date!]
_is_null - Boolean
_lt - date
_lte - date
_neq - date
_nin - [date!]
Example
{
  "_eq": date,
  "_gt": date,
  "_gte": date,
  "_in": [date],
  "_is_null": false,
  "_lt": date,
  "_lte": date,
  "_neq": date,
  "_nin": [date]
}
Types

jsonb

Example
object
Types

jsonb_comparison_exp

Boolean expression to compare columns of type "jsonb". All fields are combined with logical 'AND'.

Input Field Description
_contained_in - jsonb

is the column contained in the given json value
_contains - jsonb

does the column contain the given json value at the top level
_eq - jsonb
_gt - jsonb
_gte - jsonb
_has_key - String

does the string exist as a top-level key in the column
_has_keys_all - [String!]

do all of these strings exist as top-level keys in the column
_has_keys_any - [String!]

do any of these strings exist as top-level keys in the column
_in - [jsonb!]
_is_null - Boolean
_lt - jsonb
_lte - jsonb
_neq - jsonb
_nin - [jsonb!]
Example
{
  "_contained_in": jsonb,
  "_contains": jsonb,
  "_eq": jsonb,
  "_gt": jsonb,
  "_gte": jsonb,
  "_has_key": "abc123",
  "_has_keys_all": ["xyz789"],
  "_has_keys_any": ["xyz789"],
  "_in": [jsonb],
  "_is_null": true,
  "_lt": jsonb,
  "_lte": jsonb,
  "_neq": jsonb,
  "_nin": [jsonb]
}
Types

numeric

Example
object
Types

numeric_comparison_exp

Boolean expression to compare columns of type "numeric". All fields are combined with logical 'AND'.

Input Field Description
_eq - numeric
_gt - numeric
_gte - numeric
_in - [numeric!]
_is_null - Boolean
_lt - numeric
_lte - numeric
_neq - numeric
_nin - [numeric!]
Example
{
  "_eq": numeric,
  "_gt": numeric,
  "_gte": numeric,
  "_in": [numeric],
  "_is_null": false,
  "_lt": numeric,
  "_lte": numeric,
  "_neq": numeric,
  "_nin": [numeric]
}