Adverse Drug Event Signal Detection & Analysis

药物不良事件信号检测与分析

Automated pipeline for detecting, quantifying, and contextualizing adverse drug event signals using FAERS disproportionality analysis, FDA label mining, mechanism-based prediction, and literature evidence. Produces a quantitative Safety Signal Score (0-100) for regulatory and clinical decision-making.

KEY PRINCIPLES:

Signal quantification first - Every adverse event must have PRR/ROR/IC with confidence intervals
Serious events priority - Deaths, hospitalizations, life-threatening events always analyzed first
Multi-source triangulation - FAERS + FDA labels + OpenTargets + DrugBank + literature
Context-aware assessment - Distinguish drug-specific vs class-wide vs confounding signals
Report-first approach - Create report file FIRST, update progressively
Evidence grading mandatory - T1 (regulatory/boxed warning) through T4 (computational)
English-first queries - Always use English drug names in tool calls, respond in user's language

一条自动化流程，利用FAERS比例失衡分析、FDA说明书挖掘、基于机制的预测以及文献证据，检测、量化并关联药物不良事件信号。生成用于监管和临床决策的定量安全信号评分（0-100分）。

核心原则:

信号量化优先 - 每个不良事件必须计算带有置信区间的PRR/ROR/IC值
严重事件优先 - 死亡、住院、危及生命的事件始终优先分析
多源三角验证 - 整合FAERS + FDA说明书 + OpenTargets + DrugBank + 文献数据
上下文感知评估 - 区分药物特异性、同类药物共性及混杂信号
报告优先方法 - 先创建报告文件，再逐步更新内容
强制证据分级 - 从T1（监管/黑框警告）到T4（计算预测）的证据分级
英文优先查询 - 工具调用时始终使用英文药名，以用户语言回复

When to Use

适用场景

Apply when user asks:

"What are the safety signals for [drug]?"
"Detect adverse events for [drug]"
"Is [drug] associated with [adverse event]?"
"What are the FAERS signals for [drug]?"
"Compare safety of [drug A] vs [drug B] for [adverse event]"
"What are the serious adverse events for [drug]?"
"Are there emerging safety signals for [drug]?"
"Post-market surveillance report for [drug]"
"Pharmacovigilance signal detection for [drug]"
"What is the disproportionality analysis for [drug] and [event]?"

Differentiation from tooluniverse-pharmacovigilance: This skill focuses specifically on signal detection and quantification using disproportionality analysis (PRR, ROR, IC) with statistical rigor, produces a quantitative Safety Signal Score (0-100), and performs comparative safety analysis across drug classes. The pharmacovigilance skill provides broader safety profiling without the same depth of signal detection metrics.

当用户提出以下问题时适用:

"[药物]的安全信号有哪些？"
"检测[药物]的不良事件"
"[药物]是否与[不良事件]相关？"
"[药物]的FAERS信号有哪些？"
"比较[药物A]与[药物B]在[不良事件]上的安全性"
"[药物]的严重不良事件有哪些？"
"[药物]是否存在新出现的安全信号？"
"[药物]的上市后监测报告"
"[药物]的药物警戒信号检测"
"[药物]与[事件]的比例失衡分析结果是什么？"

与tooluniverse-pharmacovigilance的区别: 本技能专注于通过比例失衡分析（PRR、ROR、IC）进行信号检测与量化，具备统计严谨性，可生成定量安全信号评分（0-100分），并支持跨药物类别的安全性对比分析。而药物警戒技能提供更广泛的安全性分析，但不具备同等深度的信号检测指标。

Workflow Overview

工作流程概述

Phase 0: Input Parsing & Drug Disambiguation
  Parse drug name, resolve to ChEMBL ID, DrugBank ID
  Identify drug class, mechanism, and approved indications
    |
Phase 1: FAERS Adverse Event Profiling
  Top adverse events by frequency
  Seriousness and outcome distributions
  Demographics (age, sex, country)
    |
Phase 2: Disproportionality Analysis (Signal Detection)
  Calculate PRR, ROR, IC with 95% CI for each AE
  Apply signal detection criteria
  Classify signal strength (Strong/Moderate/Weak/None)
    |
Phase 3: FDA Label Safety Information
  Boxed warnings, contraindications
  Warnings and precautions, adverse reactions
  Drug interactions, special populations
    |
Phase 4: Mechanism-Based Adverse Event Context
  Target-based AE prediction (OpenTargets safety)
  Off-target effects, ADMET predictions
  Drug class effects comparison
    |
Phase 5: Comparative Safety Analysis
  Compare to drugs in same class
  Identify unique vs class-wide signals
  Head-to-head disproportionality comparison
    |
Phase 6: Drug-Drug Interactions & Risk Factors
  Known DDIs causing AEs
  Pharmacogenomic risk factors (PharmGKB)
  FDA PGx biomarkers
    |
Phase 7: Literature Evidence
  PubMed safety studies, case reports
  OpenAlex citation analysis
  Preprint emerging signals (EuropePMC)
    |
Phase 8: Risk Assessment & Safety Signal Score
  Calculate Safety Signal Score (0-100)
  Evidence grading (T1-T4) for each signal
  Clinical significance assessment
    |
Phase 9: Report Synthesis & Recommendations
  Monitoring recommendations
  Risk mitigation strategies
  Completeness checklist

Phase 0: 输入解析与药物消歧
  解析药名，关联至ChEMBL ID、DrugBank ID
  识别药物类别、作用机制及获批适应症
    |
Phase 1: FAERS不良事件分析
  按频率排序的 top 不良事件
  严重程度与结局分布
  人口统计学特征（年龄、性别、国家）
    |
Phase 2: 比例失衡分析（信号检测）
  计算每个不良事件的PRR、ROR、IC及95%置信区间
  应用信号检测标准
  分类信号强度（强/中/弱/无）
    |
Phase 3: FDA说明书安全信息
  黑框警告、禁忌症
  警告与注意事项、不良反应
  药物相互作用、特殊人群
    |
Phase 4: 基于作用机制的不良事件关联
  基于靶点的不良事件预测（OpenTargets安全数据）
  脱靶效应、ADMET预测
  同类药物效应对比
    |
Phase 5: 安全性对比分析
  与同类药物对比
  识别独特信号与同类共性信号
  头对头比例失衡对比
    |
Phase 6: 药物相互作用与风险因素
  已知可引发不良事件的药物相互作用（DDI）
  药物基因组学风险因素（PharmGKB）
  FDA PGx生物标志物
    |
Phase 7: 文献证据
  PubMed安全研究、病例报告
  OpenAlex引用分析
  EuropePMC预印本新出现信号
    |
Phase 8: 风险评估与安全信号评分
  计算安全信号评分（0-100分）
  为每个信号进行证据分级（T1-T4）
  临床意义评估
    |
Phase 9: 报告合成与建议
  监测建议
  风险缓解策略
  完整性检查清单

Phase 0: Input Parsing & Drug Disambiguation

Phase 0: 输入解析与药物消歧

0.1 Resolve Drug Identity

0.1 确认药物身份

python

undefined

python

undefined

Step 1: Get ChEMBL ID from drug name

Step 1: 通过药名获取ChEMBL ID

chembl_result = tu.tools.OpenTargets_get_drug_chembId_by_generic_name(drugName="atorvastatin")

Response: {data: {search: {hits: [{id: "CHEMBL1487", name: "ATORVASTATIN", description: "..."}]}}}

chembl_id = chembl_result['data']['search']['hits'][0]['id'] # "CHEMBL1487"

Step 2: Get drug mechanism of action

Step 2: 获取药物作用机制

moa = tu.tools.OpenTargets_get_drug_mechanisms_of_action_by_chemblId(chemblId=chembl_id)

Response: {data: {drug: {mechanismsOfAction: {rows: [{mechanismOfAction: "HMG-CoA reductase inhibitor", actionType: "INHIBITOR", targetName: "...", targets: [{id: "ENSG00000113161", approvedSymbol: "HMGCR"}]}]}}}}

Step 3: Get blackbox warning status

Step 3: 获取黑框警告状态

blackbox = tu.tools.OpenTargets_get_drug_blackbox_status_by_chembl_ID(chemblId=chembl_id)

Response: {data: {drug: {name: "ATORVASTATIN", hasBeenWithdrawn: false, blackBoxWarning: false}}}

Step 4: Get DrugBank info (safety, toxicity)

Step 4: 获取DrugBank信息（安全性、毒性）

drugbank = tu.tools.drugbank_get_safety_by_drug_name_or_drugbank_id( query="atorvastatin", case_sensitive=False, exact_match=False, limit=3 )

Response: {results: [{drug_name: "Atorvastatin", drugbank_id: "DB01076", toxicity: "...", food_interactions: "..."}]}

Step 5: Get DrugBank targets

Step 5: 获取DrugBank靶点

targets = tu.tools.drugbank_get_targets_by_drug_name_or_drugbank_id( query="atorvastatin", case_sensitive=False, exact_match=False, limit=3 )

Response: {results: [{drug_name: "...", targets: [{name: "HMG-CoA reductase", ...}]}]}

Step 6: Get approved indications

Step 6: 获取获批适应症

indications = tu.tools.OpenTargets_get_drug_indications_by_chemblId(chemblId=chembl_id)

Response: {data: {drug: {indications: {rows: [{disease: {name: "hypercholesterolemia"}, maxPhaseForIndication: 4}]}}}}

undefined

undefined

0.2 Output for Report

0.2 报告输出内容

markdown

undefined

markdown

undefined

1. Drug Identification

1. 药物识别

Property	Value
Generic Name	Atorvastatin
ChEMBL ID	CHEMBL1487
DrugBank ID	DB01076
Drug Class	HMG-CoA reductase inhibitor (Statin)
Mechanism	HMG-CoA reductase inhibitor (target: HMGCR)
Primary Target	HMGCR (ENSG00000113161)
Black Box Warning	No
Withdrawn	No

Source: OpenTargets, DrugBank

---

属性	值
通用名	Atorvastatin
ChEMBL ID	CHEMBL1487
DrugBank ID	DB01076
药物类别	HMG-CoA还原酶抑制剂（他汀类）
作用机制	HMG-CoA还原酶抑制剂（靶点：HMGCR）
主要靶点	HMGCR（ENSG00000113161）
黑框警告	无
撤市状态	未撤市

数据来源: OpenTargets, DrugBank

---

Phase 1: FAERS Adverse Event Profiling

Phase 1: FAERS不良事件分析

1.1 Query FAERS for Adverse Events

1.1 查询FAERS不良事件

python

undefined

python

undefined

Get top adverse event reactions (returns list of {term, count})

获取top不良事件反应（返回{term, count}列表）

reactions = tu.tools.FAERS_count_reactions_by_drug_event(medicinalproduct="ATORVASTATIN")

Response: [{term: "FATIGUE", count: 19171}, {term: "DIARRHOEA", count: 17127}, ...]

Get seriousness classification

获取严重程度分类

seriousness = tu.tools.FAERS_count_seriousness_by_drug_event(medicinalproduct="ATORVASTATIN")

Response: [{term: "Serious", count: 242757}, {term: "Non-serious", count: 83504}]

Get outcome distribution

获取结局分布

outcomes = tu.tools.FAERS_count_outcomes_by_drug_event(medicinalproduct="ATORVASTATIN")

Response: [{term: "Unknown", count: 162310}, {term: "Fatal", count: 22128}, ...]

Get age distribution

获取年龄分布

age_dist = tu.tools.FAERS_count_patient_age_distribution(medicinalproduct="ATORVASTATIN")

Response: [{term: "Elderly", count: 38510}, {term: "Adult", count: 24302}, ...]

Get death-related events

获取死亡相关事件

deaths = tu.tools.FAERS_count_death_related_by_drug(medicinalproduct="ATORVASTATIN")

Response: [{term: "alive", count: 113157}, {term: "death", count: 26909}]

Get reporter country distribution

获取报告者国家分布

countries = tu.tools.FAERS_count_reportercountry_by_drug_event(medicinalproduct="ATORVASTATIN")

Response: [{term: "US", count: 170963}, {term: "GB", count: 40079}, ...]

undefined

undefined

1.2 Get Serious Events Breakdown

1.2 严重事件细分

python

undefined

python

undefined

Filter serious events - all types

筛选所有类型的严重事件

serious_all = tu.tools.FAERS_filter_serious_events( operation="filter_serious_events", drug_name="ATORVASTATIN", seriousness_type="all" )

Response: {status: "success", drug_name: "ATORVASTATIN", seriousness_type: "all",

total_serious_events: N, top_serious_reactions: [{reaction: "...", count: N}, ...]}

Death-related serious events

死亡相关严重事件

serious_death = tu.tools.FAERS_filter_serious_events( operation="filter_serious_events", drug_name="ATORVASTATIN", seriousness_type="death" )

Response: {status: "success", total_serious_events: 18720,

top_serious_reactions: [{reaction: "DEATH", count: 7541}, {reaction: "MYOCARDIAL INFARCTION", count: 1286}, ...]}

Hospitalization-related

住院相关严重事件

serious_hosp = tu.tools.FAERS_filter_serious_events( operation="filter_serious_events", drug_name="ATORVASTATIN", seriousness_type="hospitalization" )

Life-threatening

危及生命的严重事件

serious_lt = tu.tools.FAERS_filter_serious_events( operation="filter_serious_events", drug_name="ATORVASTATIN", seriousness_type="life_threatening" )

undefined

serious_lt = tu.tools.FAERS_filter_serious_events( operation="filter_serious_events", drug_name="ATORVASTATIN", seriousness_type="life_threatening" )

undefined

1.3 MedDRA Hierarchy Rollup

1.3 MedDRA层级汇总

python

undefined

python

undefined

Get MedDRA preferred term rollup (top 50)

获取MedDRA首选术语汇总（top 50）

meddra = tu.tools.FAERS_rollup_meddra_hierarchy( operation="rollup_meddra_hierarchy", drug_name="ATORVASTATIN" )

Response: {status: "success", drug_name: "ATORVASTATIN",

meddra_hierarchy: {PT_level: [{preferred_term: "FATIGUE", count: 13957}, ...]}}

undefined

undefined

1.4 Output for Report

1.4 报告输出内容

markdown

undefined

markdown

undefined

2. FAERS Adverse Event Profile

2. FAERS不良事件分析

2.1 Overview

2.1 概述

Total reports: 326,261 (Serious: 242,757 | Non-serious: 83,504)
Fatal outcomes: 22,128
Primary reporter countries: US (170,963), GB (40,079), CA (16,492)

总报告数: 326,261（严重：242,757 | 非严重：83,504）
致死结局: 22,128
主要报告国家: 美国（170,963）、英国（40,079）、加拿大（16,492）

2.2 Top 10 Adverse Events by Frequency

2.2 按频率排序的Top 10不良事件

Rank	Adverse Event	Reports	% of Total
1	Fatigue	19,171	5.9%
2	Diarrhoea	17,127	5.2%
3	Dyspnoea	15,992	4.9%
...	...	...	...

排名	不良事件	报告数	占比
1	疲劳	19,171	5.9%
2	腹泻	17,127	5.2%
3	呼吸困难	15,992	4.9%
...	...	...	...

2.3 Outcome Distribution

2.3 结局分布

Outcome	Count	Percentage
Unknown	162,310	39.6%
Recovered/resolved	94,737	23.1%
Not recovered	77,721	18.9%
Recovering	49,367	12.0%
Fatal	22,128	5.4%
Recovered with sequelae	4,930	1.2%

结局	数量	占比
未知	162,310	39.6%
恢复/缓解	94,737	23.1%
未恢复	77,721	18.9%
恢复中	49,367	12.0%
致死	22,128	5.4%
遗留后遗症的恢复	4,930	1.2%

2.4 Age Distribution

2.4 年龄分布

Age Group	Reports	Percentage
Elderly	38,510	61.3%
Adult	24,302	38.7%
Other	152	<1%

Source: FAERS via FAERS_count_reactions_by_drug_event, FAERS_count_seriousness_by_drug_event

---

年龄组	报告数	占比
老年	38,510	61.3%
成年	24,302	38.7%
其他	152	<1%

数据来源: FAERS via FAERS_count_reactions_by_drug_event, FAERS_count_seriousness_by_drug_event

---

Phase 2: Disproportionality Analysis (Signal Detection)

Phase 2: 比例失衡分析（信号检测）

2.1 Calculate Signal Metrics

2.1 计算信号指标

CRITICAL: This is the core of the skill. For each top adverse event (at least top 15-20), calculate PRR, ROR, and IC with 95% confidence intervals.

python

undefined

核心步骤: 本技能的核心环节。针对每个top不良事件（至少top 15-20），计算PRR、ROR、IC及95%置信区间。

python

undefined

For each significant adverse event, calculate disproportionality

针对每个显著不良事件，计算比例失衡指标

top_events = ["Rhabdomyolysis", "Myalgia", "Hepatotoxicity", "Diabetes mellitus", "Acute kidney injury", "Myopathy", "Pancreatitis"]

for event in top_events: result = tu.tools.FAERS_calculate_disproportionality( operation="calculate_disproportionality", drug_name="ATORVASTATIN", adverse_event=event ) # Response structure: # { # status: "success", # drug_name: "ATORVASTATIN", # adverse_event: "Rhabdomyolysis", # contingency_table: { # a_drug_and_event: 2226, # b_drug_no_event: 241655, # c_no_drug_event: 37658, # d_no_drug_no_event: 19725450 # }, # metrics: { # ROR: {value: 4.825, ci_95_lower: 4.622, ci_95_upper: 5.037}, # PRR: {value: 4.79, ci_95_lower: 4.59, ci_95_upper: 4.998}, # IC: {value: 2.194, ci_95_lower: 2.136, ci_95_upper: 2.252} # }, # signal_detection: { # signal_detected: true, # signal_strength: "Strong signal", # criteria: "ROR lower CI > 1.0 and case count >= 3" # } # }

undefined

top_events = ["Rhabdomyolysis", "Myalgia", "Hepatotoxicity", "Diabetes mellitus", "Acute kidney injury", "Myopathy", "Pancreatitis"]

for event in top_events: result = tu.tools.FAERS_calculate_disproportionality( operation="calculate_disproportionality", drug_name="ATORVASTATIN", adverse_event=event ) # Response结构: # { # status: "success", # drug_name: "ATORVASTATIN", # adverse_event: "Rhabdomyolysis", # contingency_table: { # a_drug_and_event: 2226, # b_drug_no_event: 241655, # c_no_drug_event: 37658, # d_no_drug_no_event: 19725450 # }, # metrics: { # ROR: {value: 4.825, ci_95_lower: 4.622, ci_95_upper: 5.037}, # PRR: {value: 4.79, ci_95_lower: 4.59, ci_95_upper: 4.998}, # IC: {value: 2.194, ci_95_lower: 2.136, ci_95_upper: 2.252} # }, # signal_detection: { # signal_detected: true, # signal_strength: "Strong signal", # criteria: "ROR lower CI > 1.0 and case count >= 3" # } # }

undefined

2.2 Signal Detection Criteria

2.2 信号检测标准

Proportional Reporting Ratio (PRR):

PRR = (a/(a+b)) / (c/(c+d))
Signal: PRR >= 2.0 AND lower 95% CI > 1.0 AND case count >= 3

Reporting Odds Ratio (ROR):

ROR = (ad) / (bc)
Signal: Lower 95% CI > 1.0

Information Component (IC):

IC = log2(observed/expected)
Signal: Lower 95% CI > 0

比例报告比（PRR）:

PRR = (a/(a+b)) / (c/(c+d))
信号判定: PRR >= 2.0 且 95%置信区间下限 > 1.0 且病例数 >= 3

报告比值比（ROR）:

ROR = (ad) / (bc)
信号判定: 95%置信区间下限 > 1.0

信息成分（IC）:

IC = log2(实际值/预期值)
信号判定: 95%置信区间下限 > 0

2.3 Signal Strength Classification

2.3 信号强度分类

Strength	PRR	ROR Lower CI	IC Lower CI	Clinical Action
Strong	>= 5.0	>= 3.0	>= 2.0	Immediate investigation required
Moderate	3.0-4.9	2.0-2.9	1.0-1.9	Active monitoring recommended
Weak	2.0-2.9	1.0-1.9	0-0.9	Routine monitoring, watch for trends
No signal	< 2.0	< 1.0	< 0	Standard pharmacovigilance

强度	PRR	ROR 95%置信区间下限	IC 95%置信区间下限	临床行动
强	>= 5.0	>= 3.0	>= 2.0	需立即开展调查
中	3.0-4.9	2.0-2.9	1.0-1.9	建议主动监测
弱	2.0-2.9	1.0-1.9	0-0.9	常规监测，关注趋势
无信号	< 2.0	< 1.0	< 0	标准药物警戒流程

2.4 Demographic Stratification of Key Signals

2.4 关键信号的人口统计学分层

python

undefined

python

undefined

For strong/moderate signals, stratify by demographics

针对强/中度信号，按人口统计学特征分层

result = tu.tools.FAERS_stratify_by_demographics( operation="stratify_by_demographics", drug_name="ATORVASTATIN", adverse_event="Rhabdomyolysis", stratify_by="sex" # Options: sex, age, country )

result = tu.tools.FAERS_stratify_by_demographics( operation="stratify_by_demographics", drug_name="ATORVASTATIN", adverse_event="Rhabdomyolysis", stratify_by="sex" # 可选值: sex, age, country )

Response: {status: "success", total_reports: 1996,

stratification: [{group: 1, count: 1190, percentage: 59.62}, # 1=Male

stratification: [{group: 1, count: 1190, percentage: 59.62}, # 1=男性

{group: 2, count: 781, percentage: 39.13}]} # 2=Female

{group: 2, count: 781, percentage: 39.13}]} # 2=女性


**Note on sex codes**: group 0 = Unknown, group 1 = Male, group 2 = Female.


**性别编码说明**: group 0 = 未知, group 1 = 男性, group 2 = 女性。

2.5 Output for Report

2.5 报告输出内容

markdown

undefined

markdown

undefined

3. Disproportionality Analysis (Signal Detection)

3. 比例失衡分析（信号检测）

3.1 Signal Detection Summary

3.1 信号检测汇总

Adverse Event	Cases (a)	PRR	PRR 95% CI	ROR	ROR 95% CI	IC	Signal
Rhabdomyolysis	2,226	4.79	4.59-5.00	4.83	4.62-5.04	2.19	STRONG
Myopathy	1,234	6.12	5.72-6.55	6.18	5.77-6.62	2.54	STRONG
Myalgia	9,189	2.31	2.26-2.37	2.33	2.28-2.39	1.18	Moderate
Hepatotoxicity	456	3.45	3.10-3.84	3.48	3.13-3.87	1.72	Moderate
Diabetes mellitus	3,021	1.89	1.82-1.96	1.90	1.83-1.97	0.91	Weak
Pancreatitis	678	2.15	1.97-2.34	2.16	1.98-2.35	1.08	Weak

不良事件	病例数（a）	PRR	PRR 95%置信区间	ROR	ROR 95%置信区间	IC	信号强度
横纹肌溶解症	2,226	4.79	4.59-5.00	4.83	4.62-5.04	2.19	强
肌病	1,234	6.12	5.72-6.55	6.18	5.77-6.62	2.54	强
肌痛	9,189	2.31	2.26-2.37	2.33	2.28-2.39	1.18	中
肝毒性	456	3.45	3.10-3.84	3.48	3.13-3.87	1.72	中
糖尿病	3,021	1.89	1.82-1.96	1.90	1.83-1.97	0.91	弱
胰腺炎	678	2.15	1.97-2.34	2.16	1.98-2.35	1.08	弱

3.2 Demographics of Key Signals

3.2 关键信号的人口统计学特征

Rhabdomyolysis (n=1,996):

Male: 59.6%, Female: 39.1%, Unknown: 1.3%
Predominantly elderly (>65 years)

Source: FAERS via FAERS_calculate_disproportionality, FAERS_stratify_by_demographics

---

横纹肌溶解症（n=1,996）:

男性: 59.6%, 女性: 39.1%, 未知: 1.3%
主要集中于老年人群（>65岁）

数据来源: FAERS via FAERS_calculate_disproportionality, FAERS_stratify_by_demographics

---

Phase 3: FDA Label Safety Information

Phase 3: FDA说明书安全信息

3.1 Extract Label Sections

3.1 提取说明书章节

python

undefined

python

undefined

Boxed warnings

黑框警告

boxed = tu.tools.FDA_get_boxed_warning_info_by_drug_name(drug_name="atorvastatin")

Response: {meta: {total: N}, results: [{boxed_warning: ["...text..."]}]}

NOTE: Returns {error: {code: "NOT_FOUND"}} if no boxed warning exists

注意: 若无黑框警告，返回{error: {code: "NOT_FOUND"}}

Contraindications

禁忌症

contras = tu.tools.FDA_get_contraindications_by_drug_name(drug_name="atorvastatin")

Response: {meta: {total: N}, results: [{openfda.generic_name: [...], contraindications: ["...text..."]}]}

Warnings and precautions

警告与注意事项

warnings = tu.tools.FDA_get_warnings_by_drug_name(drug_name="atorvastatin")

Response: {meta: {total: N}, results: [{warnings: ["...text..."], boxed_warning: [...]}]}

Adverse reactions from label

说明书中的不良反应

adverse_rxns = tu.tools.FDA_get_adverse_reactions_by_drug_name(drug_name="atorvastatin")

Response: {meta: {total: N}, results: [{adverse_reactions: ["...text..."]}]}

Drug interactions from label

说明书中的药物相互作用

interactions = tu.tools.FDA_get_drug_interactions_by_drug_name(drug_name="atorvastatin")

Response: {meta: {total: N}, results: [{drug_interactions: ["...text..."]}]}

Pregnancy/breastfeeding

妊娠/哺乳期信息

pregnancy = tu.tools.FDA_get_pregnancy_or_breastfeeding_info_by_drug_name(drug_name="atorvastatin")

Geriatric use

老年用药

geriatric = tu.tools.FDA_get_geriatric_use_info_by_drug_name(drug_name="atorvastatin")

Pediatric use

儿科用药

pediatric = tu.tools.FDA_get_pediatric_use_info_by_drug_name(drug_name="atorvastatin")

Pharmacogenomics from label

说明书中的药物基因组学信息

pgx_label = tu.tools.FDA_get_pharmacogenomics_info_by_drug_name(drug_name="atorvastatin")

undefined

pgx_label = tu.tools.FDA_get_pharmacogenomics_info_by_drug_name(drug_name="atorvastatin")

undefined

3.2 Handling No Results

3.2 无结果处理

IMPORTANT: FDA label tools return

{error: {code: "NOT_FOUND"}}

when a section does not exist. This is NORMAL for many drugs - for example, most drugs do NOT have boxed warnings. Always check for this pattern:

python

undefined

重要提示: FDA说明书工具在对应章节不存在时，会返回

{error: {code: "NOT_FOUND"}}

。这在多数药物中属于正常情况——例如，多数药物无黑框警告。需始终检查以下模式:

python

undefined

Check if boxed warning exists

检查是否存在黑框警告

if isinstance(boxed, dict) and 'error' in boxed: boxed_warning_text = "None (no boxed warning for this drug)" else: boxed_warning_text = boxed['results'][0].get('boxed_warning', ['None'])[0]

undefined

if isinstance(boxed, dict) and 'error' in boxed: boxed_warning_text = "无（本药物无黑框警告）" else: boxed_warning_text = boxed['results'][0].get('boxed_warning', ['无'])[0]

undefined

3.3 Output for Report

3.3 报告输出内容

markdown

undefined

markdown

undefined

4. FDA Label Safety Information

4. FDA说明书安全信息

4.1 Boxed Warning

4.1 黑框警告

None

无

4.2 Contraindications

4.2 禁忌症

Acute liver failure or decompensated cirrhosis
Hypersensitivity to atorvastatin (includes anaphylaxis, angioedema, SJS, TEN)

急性肝衰竭或失代偿期肝硬化
对阿托伐他汀过敏（包括过敏反应、血管性水肿、Stevens-Johnson综合征、中毒性表皮坏死症）

4.3 Warnings and Precautions

4.3 警告与注意事项

Warning	Clinical Relevance
Myopathy/Rhabdomyolysis	Risk with CYP3A4 inhibitors, high doses
Immune-Mediated Necrotizing Myopathy	Rare autoimmune myopathy
Hepatic Dysfunction	Monitor LFTs
Increased HbA1c/Glucose	Diabetes risk

警告	临床相关性
肌病/横纹肌溶解症	与CYP3A4抑制剂、高剂量合用时风险升高
免疫介导性坏死性肌病	罕见自身免疫性肌病
肝功能异常	需监测肝功能指标（LFTs）
HbA1c/血糖升高	糖尿病风险

4.4 Drug Interactions (from label)

4.4 说明书中的药物相互作用

Interacting Drug	Mechanism	Clinical Action
Cyclosporine	Increased exposure	Avoid combination
CYP3A4 inhibitors	Increased atorvastatin levels	Use lowest dose
Gemfibrozil	Increased myopathy risk	Avoid

相互作用药物	作用机制	临床行动
环孢素	暴露量升高	避免合用
CYP3A4抑制剂	阿托伐他汀水平升高	使用最低有效剂量
吉非贝齐	肌病风险升高	避免合用

4.5 Special Populations

4.5 特殊人群

Pregnancy: Contraindicated
Geriatric: No dose adjustment needed
Pediatric: Approved for heterozygous FH ages 10+

Source: FDA drug labels via FDA_get_contraindications_by_drug_name, FDA_get_warnings_by_drug_name

---

妊娠: 禁用
老年: 无需调整剂量
儿科: 获批用于10岁以上杂合子型家族性高胆固醇血症患者

数据来源: FDA药品说明书 via FDA_get_contraindications_by_drug_name, FDA_get_warnings_by_drug_name

---

Phase 4: Mechanism-Based Adverse Event Context

Phase 4: 基于作用机制的不良事件关联

4.1 Target Safety Profile

4.1 靶点安全特征

python

undefined

python

undefined

Get target safety data from OpenTargets

从OpenTargets获取靶点安全数据

First get target ensembl ID from MOA result

先从Phase 0的作用机制结果中获取靶点ensembl ID

target_id = "ENSG00000113161" # HMGCR from Phase 0

safety = tu.tools.OpenTargets_get_target_safety_profile_by_ensemblID(ensemblId=target_id)

target_id = "ENSG00000113161" # HMGCR（来自Phase 0）

safety = tu.tools.OpenTargets_get_target_safety_profile_by_ensemblID(ensemblId=target_id)

Response: {data: {target: {id: "...", approvedSymbol: "HMGCR",

safetyLiabilities: [{event: "Decrease, Fertility", eventId: "...",

effects: [{direction: "Inhibition/Decrease/Downregulation"}],

studies: [{type: "cell-based"}], datasource: "AOP-Wiki"}]}}}

Get OpenTargets adverse events (uses FAERS data)

获取OpenTargets不良事件（基于FAERS数据）

ot_aes = tu.tools.OpenTargets_get_drug_adverse_events_by_chemblId(chemblId="CHEMBL1487")

Response: {data: {drug: {adverseEvents: {count: 13, criticalValue: 513.67,

rows: [{name: "myalgia", meddraCode: "10028411", count: 4126, logLR: 6067.33}, ...]}}}}

undefined

undefined

4.2 ADMET Predictions (if SMILES available)

4.2 ADMET预测（若SMILES可用）

python

undefined

python

undefined

Get SMILES from DrugBank/PharmGKB

从DrugBank/PharmGKB获取SMILES

smiles = "CC(C)C1=C(C(=C(N1CCC@HO)C2=CC=C(C=C2)F)C3=CC=CC=C3)C(=O)NC4=CC=CC=C4"

Toxicity predictions

毒性预测

toxicity = tu.tools.ADMETAI_predict_toxicity(smiles=[smiles])

Response: predictions for hepatotoxicity, cardiotoxicity, etc.

Response: 肝毒性、心脏毒性等预测结果

CYP interaction predictions

CYP相互作用预测

cyp = tu.tools.ADMETAI_predict_CYP_interactions(smiles=[smiles])

Response: CYP inhibition/substrate predictions

Response: CYP抑制/底物预测结果

undefined

undefined

4.3 Drug Warnings from OpenTargets

4.3 OpenTargets药物警告

python

undefined

python

undefined

Drug warnings (withdrawals, safety warnings)

药物警告（撤市、安全警告）

warnings = tu.tools.OpenTargets_get_drug_warnings_by_chemblId(chemblId="CHEMBL1487")

Response: {data: {drug: {id: "CHEMBL1487", name: "ATORVASTATIN"}}}

Note: Empty if no warnings exist

注意: 若无警告则返回空

undefined

undefined

4.4 Output for Report

4.4 报告输出内容

markdown

undefined

markdown

undefined

5. Mechanism-Based Adverse Event Context

5. 基于作用机制的不良事件关联

5.1 Target Safety Profile (HMGCR)

5.1 靶点安全特征（HMGCR）

Safety Liability	Direction	Evidence	Source
Decreased fertility	Inhibition	Cell-based	AOP-Wiki

安全风险	作用方向	证据类型	数据来源
生育能力下降	抑制	细胞实验	AOP-Wiki

5.2 OpenTargets Significant Adverse Events

5.2 OpenTargets显著不良事件

Adverse Event	FAERS Count	log(LR)	MedDRA Code
Myalgia	4,126	6,067	10028411
Rhabdomyolysis	2,546	4,788	10039020
CPK increased	1,709	2,909	10005470

不良事件	FAERS报告数	log(LR)	MedDRA编码
肌痛	4,126	6,067	10028411
横纹肌溶解症	2,546	4,788	10039020
CPK升高	1,709	2,909	10005470

5.3 ADMET Predictions

5.3 ADMET预测

Property	Prediction	Confidence
Hepatotoxicity	Moderate risk	0.65
Cardiotoxicity (hERG)	Low risk	0.23
CYP3A4 substrate	Yes	0.92

Source: OpenTargets, ADMETAI

---

属性	预测结果	置信度
肝毒性	中度风险	0.65
心脏毒性（hERG）	低风险	0.23
CYP3A4底物	是	0.92

数据来源: OpenTargets, ADMETAI

---

Phase 5: Comparative Safety Analysis

Phase 5: 安全性对比分析

5.1 Compare to Drug Class

5.1 与同类药物对比

python

undefined

python

undefined

Head-to-head comparison with class member

与同类药物头对头对比

comparison = tu.tools.FAERS_compare_drugs( operation="compare_drugs", drug1="ATORVASTATIN", drug2="SIMVASTATIN", adverse_event="Rhabdomyolysis" )

Response: {status: "success", adverse_event: "Rhabdomyolysis",

drug1: {name: "ATORVASTATIN", metrics: {PRR: {value: 4.79, ...}, ROR: {...}, IC: {...}},

signal_detection: {signal_detected: true, signal_strength: "Strong signal"}},

drug2: {name: "SIMVASTATIN", metrics: {PRR: {value: 12.78, ...}, ...}},

comparison: "SIMVASTATIN shows stronger signal than ATORVASTATIN"}

Compare multiple events across class members

对比同类药物的多个事件

class_drugs = ["ATORVASTATIN", "SIMVASTATIN", "ROSUVASTATIN", "PRAVASTATIN"] key_events = ["Rhabdomyolysis", "Myalgia", "Hepatotoxicity", "Diabetes mellitus"]

Run FAERS_compare_drugs for each pair and event combination

为每对药物和事件组合运行FAERS_compare_drugs

Aggregate adverse events across drug class

汇总同类药物的不良事件

class_aes = tu.tools.FAERS_count_additive_adverse_reactions( medicinalproducts=class_drugs )

Response: [{term: "FATIGUE", count: N}, ...]

Aggregate seriousness across class

汇总同类药物的严重程度

class_serious = tu.tools.FAERS_count_additive_seriousness_classification( medicinalproducts=class_drugs )

Response: [{term: "Serious", count: N}, {term: "Non-serious", count: N}]

undefined

undefined

5.2 Output for Report

5.2 报告输出内容

markdown

undefined

markdown

undefined

6. Comparative Safety Analysis (Statin Class)

6. 安全性对比分析（他汀类）

6.1 Head-to-Head: Rhabdomyolysis

6.1 头对头对比：横纹肌溶解症

Drug	PRR	PRR 95% CI	ROR	Cases	Signal
Simvastatin	12.78	12.43-13.14	13.05	5,234	STRONG
Atorvastatin	4.79	4.59-5.00	4.83	2,226	STRONG
Rosuvastatin	3.45	3.21-3.71	3.47	1,102	Moderate
Pravastatin	5.67	5.28-6.09	5.72	1,876	STRONG

药物	PRR	PRR 95%置信区间	ROR	病例数	信号强度
辛伐他汀	12.78	12.43-13.14	13.05	5,234	强
阿托伐他汀	4.79	4.59-5.00	4.83	2,226	强
瑞舒伐他汀	3.45	3.21-3.71	3.47	1,102	中
普伐他汀	5.67	5.28-6.09	5.72	1,876	强

6.2 Class-Wide vs Drug-Specific Signals

6.2 同类共性信号 vs 药物特异性信号

Signal Type	Events
Class-wide (all statins)	Myalgia, Rhabdomyolysis, CPK elevation, Hepatotoxicity
Drug-specific (atorvastatin)	[None identified - all signals are class-wide]

Source: FAERS via FAERS_compare_drugs

---

信号类型	事件
同类共性（所有他汀类）	肌痛、横纹肌溶解症、CPK升高、肝毒性
药物特异性（阿托伐他汀）	[未识别到 - 所有信号均为同类共性]

数据来源: FAERS via FAERS_compare_drugs

---

Phase 6: Drug-Drug Interactions & Risk Factors

Phase 6: 药物相互作用与风险因素

6.1 Drug-Drug Interactions

6.1 药物相互作用

python

undefined

python

undefined

FDA label DDIs

FDA说明书中的药物相互作用

ddi_label = tu.tools.FDA_get_drug_interactions_by_drug_name(drug_name="atorvastatin")

Response: {results: [{drug_interactions: ["...text..."]}]}

DrugBank interactions

DrugBank中的药物相互作用

ddi_db = tu.tools.drugbank_get_drug_interactions_by_drug_name_or_id( query="atorvastatin", case_sensitive=False, exact_match=False, limit=3 )

DailyMed DDIs

DailyMed中的药物相互作用

ddi_dailymed = tu.tools.DailyMed_parse_drug_interactions(drug_name="atorvastatin")

undefined

ddi_dailymed = tu.tools.DailyMed_parse_drug_interactions(drug_name="atorvastatin")

undefined

6.2 Pharmacogenomic Risk Factors

6.2 药物基因组学风险因素

python

undefined

python

undefined

PharmGKB drug search

PharmGKB药物搜索

pgx_search = tu.tools.PharmGKB_search_drugs(query="atorvastatin")

Response: {status: "success", data: [{id: "PA448500", name: "atorvastatin", smiles: "..."}]}

Get detailed PGx info

获取详细PGx信息

pgx_details = tu.tools.PharmGKB_get_drug_details(drug_id="PA448500")

PharmGKB dosing guidelines

PharmGKB给药指南

dosing = tu.tools.PharmGKB_get_dosing_guidelines(gene="SLCO1B1")

SLCO1B1 is key pharmacogene for statins

SLCO1B1是他汀类药物的关键药物基因

FDA PGx biomarkers

FDA PGx生物标志物

fda_pgx = tu.tools.fda_pharmacogenomic_biomarkers(drug_name="atorvastatin", limit=10)

Response: {count: N, results: [{drug_name: "...", biomarker: "...", ...}]}

Note: May return empty results for some drugs

注意: 部分药物可能返回空结果

undefined

undefined

6.3 Output for Report

6.3 报告输出内容

markdown

undefined

markdown

undefined

7. Drug-Drug Interactions & Pharmacogenomic Risk

7. 药物相互作用与药物基因组学风险

7.1 Key Drug-Drug Interactions

7.1 关键药物相互作用

Interacting Drug	Mechanism	Risk	Management
Cyclosporine	CYP3A4 inhibition	Rhabdomyolysis	Avoid combination
Clarithromycin	CYP3A4 inhibition	Myopathy	Limit to 20mg/day
Gemfibrozil	Glucuronidation inhibition	Myopathy	Avoid combination
Niacin (>1g/day)	Additive myotoxicity	Myopathy	Monitor closely

相互作用药物	作用机制	风险	管理策略
环孢素	CYP3A4抑制	横纹肌溶解症	避免合用
克拉霉素	CYP3A4抑制	肌病	剂量限制为20mg/日
吉非贝齐	葡萄糖醛酸化抑制	肌病	避免合用
烟酸（>1g/日）	叠加肌毒性	肌病	密切监测

7.2 Pharmacogenomic Risk Factors

7.2 药物基因组学风险因素

Gene	Variant	Phenotype	Recommendation	Evidence
SLCO1B1	rs4149056 (*5)	Reduced transport	Consider lower dose	Level 1A
CYP3A4	*22 (rs35599367)	Poor metabolizer	Increased exposure	Level 3

Source: FDA label, PharmGKB, fda_pharmacogenomic_biomarkers

---

基因	变异体	表型	建议	证据等级
SLCO1B1	rs4149056 (*5)	转运能力下降	考虑降低剂量	Level 1A
CYP3A4	*22 (rs35599367)	弱代谢型	暴露量升高	Level 3

数据来源: FDA说明书, PharmGKB, fda_pharmacogenomic_biomarkers

---

Phase 7: Literature Evidence

Phase 7: 文献证据

7.1 Search Published Literature

7.Search Published Literature

python

undefined

python

undefined

PubMed safety studies

PubMed安全研究

pubmed = tu.tools.PubMed_search_articles( query='atorvastatin adverse events safety rhabdomyolysis', limit=20 )

Response: [{pmid: "...", title: "...", authors: [...], journal: "...",

pub_date: "...", pub_year: "...", doi: "..."}]

Citation analysis via OpenAlex

OpenAlex引用分析

openalex = tu.tools.openalex_search_works( query="atorvastatin safety adverse events", limit=15 )

Preprints via EuropePMC

EuropePMC预印本

preprints = tu.tools.EuropePMC_search_articles( query="atorvastatin safety signal", source="PPR", pageSize=10 )

undefined

preprints = tu.tools.EuropePMC_search_articles( query="atorvastatin safety signal", source="PPR", pageSize=10 )

undefined

7.2 Output for Report

7.2 报告输出内容

markdown

undefined

markdown

undefined

8. Literature Evidence

8. 文献证据

8.1 Key Safety Publications

8.1 关键安全出版物

PMID	Title	Year	Journal
41657777	Differential musculoskeletal outcome reporting...	2026	Front Pharmacol
...	...	...	...

PMID	标题	年份	期刊
41657777	Differential musculoskeletal outcome reporting...	2026	Front Pharmacol
...	...	...	...

8.2 Evidence Summary

8.2 证据汇总

Evidence Type	Count	Key Findings
Meta-analyses	5	Statin myopathy 5-10%, rhabdomyolysis rare
RCTs	12	CV benefit outweighs muscle risk
Case reports	23	Severe rhabdomyolysis with CYP3A4 inhibitors

Source: PubMed, OpenAlex

---

证据类型	数量	关键发现
荟萃分析	5	他汀类肌病发生率5-10%，横纹肌溶解症罕见
随机对照试验（RCT）	12	心血管获益超过肌肉风险
病例报告	23	与CYP3A4抑制剂合用时出现严重横纹肌溶解症

数据来源: PubMed, OpenAlex

---

Phase 8: Risk Assessment & Safety Signal Score

Phase 8: 风险评估与安全信号评分

8.1 Safety Signal Score Calculation (0-100)

8.1 安全信号评分计算（0-100分）

The Safety Signal Score quantifies overall drug safety concern on a 0-100 scale (higher = more concern).

Component 1: FAERS Signal Strength (0-35 points)

If any signal has PRR >= 5 AND ROR lower CI >= 3: 35 points
If any signal has PRR 3-5 AND ROR lower CI 2-3: 20 points
If any signal has PRR 2-3 AND ROR lower CI 1-2: 10 points
If no signals detected: 0 points

Component 2: Serious Adverse Events (0-30 points)

Deaths reported with high count (>100): 30 points
Deaths reported with low count (1-100): 25 points
Life-threatening events: 20 points
Hospitalizations only: 15 points
Non-serious only: 0 points

Component 3: FDA Label Warnings (0-25 points)

Boxed warning present: 25 points
Drug withdrawn or restricted: 25 points
Contraindications present: 15 points
Warnings and precautions: 10 points
Adverse reactions only: 5 points
No label warnings: 0 points

Component 4: Literature Evidence (0-10 points)

Meta-analyses confirming safety signals: 10 points
Multiple RCTs with safety concerns: 7 points
Case reports/case series: 4 points
No published safety concerns: 0 points

Total Score Interpretation:

Score Range	Interpretation	Action
75-100	High concern	Serious safety signals; requires immediate regulatory attention
50-74	Moderate concern	Significant monitoring needed; consider risk mitigation
25-49	Low-moderate concern	Routine enhanced monitoring; standard risk management
0-24	Low concern	Standard safety profile; routine pharmacovigilance

安全信号评分以0-100分量化药物整体安全风险（分数越高，风险越高）。

成分1: FAERS信号强度（0-35分）

若存在PRR >=5且ROR置信区间下限>=3的信号: 35分
若存在PRR 3-5且ROR置信区间下限2-3的信号: 20分
若存在PRR 2-3且ROR置信区间下限1-2的信号: 10分
无信号: 0分

成分2: 严重不良事件（0-30分）

死亡报告数多（>100例）: 30分
死亡报告数少（1-100例）: 25分
存在危及生命事件: 20分
仅存在住院事件: 15分
仅存在非严重事件: 0分

成分3: FDA说明书警告（0-25分）

存在黑框警告: 25分
药物撤市或受限: 25分
存在禁忌症: 15分
存在警告与注意事项: 10分
仅存在不良反应: 5分
无说明书警告: 0分

成分4: 文献证据（0-10分）

荟萃分析确认安全信号: 10分
多项RCT提示安全风险: 7分
病例报告/病例系列: 4分
无已发表安全风险: 0分

总分解读:

分数范围	解读	行动
75-100	高风险	存在严重安全信号；需立即引起监管关注
50-74	中风险	需重点监测；考虑风险缓解措施
25-49	低-中风险	常规强化监测；标准风险管理
0-24	低风险	标准安全特征；常规药物警戒流程

8.2 Evidence Grading

8.2 证据分级

Tier	Criteria	Example
T1	Boxed warning, confirmed by RCTs, PRR > 10	Metformin: Lactic acidosis
T2	Label warning + FAERS signal (PRR 3-10) + published studies	Atorvastatin: Rhabdomyolysis
T3	FAERS signal (PRR 2-3) + case reports	Atorvastatin: Pancreatitis
T4	Computational prediction only (ADMET) or weak signal	ADMETAI hepatotoxicity prediction

等级	标准	示例
T1	黑框警告，经RCT确认，PRR>10	二甲双胍：乳酸酸中毒
T2	说明书警告 + FAERS信号（PRR 3-10） + 已发表研究	阿托伐他汀：横纹肌溶解症
T3	FAERS信号（PRR 2-3） + 病例报告	阿托伐他汀：胰腺炎
T4	仅计算预测（ADMET）或弱信号	ADMETAI肝毒性预测

8.3 Output for Report

8.3 报告输出内容

markdown

undefined

markdown

undefined

9. Risk Assessment

9. 风险评估

9.1 Safety Signal Score: 62/100 (MODERATE CONCERN)

9.1 安全信号评分: 62/100（中风险）

Component	Score	Max	Rationale
FAERS Signal Strength	35	35	Strong signals (PRR >= 5 for rhabdomyolysis)
Serious Adverse Events	15	30	Hospitalizations; deaths uncommon for drug itself
FDA Label Warnings	10	25	Warnings/precautions but no boxed warning
Literature Evidence	7	10	Multiple RCTs confirm muscle-related risks
TOTAL	62	100	MODERATE CONCERN

成分	得分	满分	依据
FAERS信号强度	35	35	存在强信号（横纹肌溶解症PRR>=5）
严重不良事件	15	30	存在住院事件；药物本身相关死亡罕见
FDA说明书警告	10	25	存在警告与注意事项，但无黑框警告
文献证据	7	10	多项RCT确认肌肉相关风险
总分	62	100	中风险

9.2 Evidence-Graded Signals

9.2 证据分级信号

Signal	Grade	PRR	Serious	Label	Literature	Overall
Rhabdomyolysis	T2	4.79	Yes	Warning	Confirmed	Moderate
Myopathy	T2	6.12	Yes	Warning	Confirmed	Moderate
Hepatotoxicity	T3	3.45	Rare	Warning	Case reports	Low-Moderate
Diabetes risk	T3	1.89	No	Warning	RCT data	Low

---

信号	等级	PRR	严重程度	说明书提及	文献支持	整体风险
横纹肌溶解症	T2	4.79	是	警告	已确认	中
肌病	T2	6.12	是	警告	已确认	中
肝毒性	T3	3.45	罕见	警告	病例报告	低-中
糖尿病风险	T3	1.89	否	警告	RCT数据	低

---

Phase 9: Report Synthesis & Recommendations

Phase 9: 报告合成与建议

9.1 Report Template

9.1 报告模板

File:

[DRUG]_adverse_event_report.md

markdown

undefined

文件:

[DRUG]_adverse_event_report.md

markdown

undefined

Adverse Drug Event Signal Detection Report: [DRUG]

药物不良事件信号检测报告: [DRUG]

Generated: [Date] | Drug: [Generic Name] | ChEMBL ID: [ID] Safety Signal Score: [XX/100] ([INTERPRETATION])

生成日期: [Date] | 药物: [通用名] | ChEMBL ID: [ID] 安全信号评分: [XX/100] ([解读])

Executive Summary

执行摘要

[2-3 paragraph summary of key findings]

Key Safety Signals:

[Strongest signal with PRR/ROR]
[Second signal]
[Third signal]

Regulatory Status: [Boxed warning Y/N] | [Withdrawn Y/N] | [Restrictions]

[2-3段关键发现总结]

关键安全信号:

[最强信号及PRR/ROR值]
[次强信号]
[第三信号]

监管状态: [是否有黑框警告] | [是否撤市] | [限制条件]

1. Drug Identification

1. 药物识别

[Phase 0 output]

[Phase 0输出内容]

2. FAERS Adverse Event Profile

2. FAERS不良事件分析

[Phase 1 output]

[Phase 1输出内容]

3. Disproportionality Analysis

3. 比例失衡分析

[Phase 2 output]

[Phase 2输出内容]

4. FDA Label Safety Information

4. FDA说明书安全信息

[Phase 3 output]

[Phase 3输出内容]

5. Mechanism-Based Context

5. 基于作用机制的关联分析

[Phase 4 output]

[Phase 4输出内容]

6. Comparative Safety Analysis

6. 安全性对比分析

[Phase 5 output]

[Phase 5输出内容]

7. Drug-Drug Interactions & PGx Risk

7. 药物相互作用与PGx风险

[Phase 6 output]

[Phase 6输出内容]

8. Literature Evidence

8. 文献证据

[Phase 7 output]

[Phase 7输出内容]

9. Risk Assessment

9. 风险评估

[Phase 8 output]

[Phase 8输出内容]

10. Clinical Recommendations

10. 临床建议

10.1 Monitoring Recommendations

10.1 监测建议

Parameter	Frequency	Rationale
[Lab test]	[Frequency]	[Why]

参数	频率	依据
[实验室检查]	[频率]	[原因]

10.2 Risk Mitigation Strategies

10.2 风险缓解策略

Risk	Mitigation	Evidence
[Risk]	[Strategy]	[Source]

风险	缓解措施	证据
[风险]	[策略]	[来源]

10.3 Patient Counseling Points

10.3 患者宣教要点

[Point 1]
[Point 2]

[要点1]
[要点2]

10.4 Populations at Higher Risk

10.4 高风险人群

Population	Risk Factor	Recommendation
[Group]	[Factor]	[Action]

人群	风险因素	建议
[群体]	[因素]	[行动]

11. Completeness Checklist

11. 完整性检查清单

[See below]

[见下文]

12. Data Sources

12. 数据来源

[All tools and databases used with timestamps]

---

[所有使用的工具和数据库及时间戳]

---

Completeness Checklist

完整性检查清单

Phase 0: Drug Disambiguation

Phase 0: 药物消歧

Phase 1: FAERS Profiling

Phase 1: FAERS分析

Top adverse events queried (>=15 events)
Seriousness distribution obtained
Outcome distribution obtained
Age distribution obtained
Death-related events counted
Reporter country distribution obtained

已查询top不良事件（>=15个事件）
已获取严重程度分布
已获取结局分布
已获取年龄分布
已统计死亡相关事件
已获取报告者国家分布

Phase 2: Disproportionality Analysis

Phase 2: 比例失衡分析

PRR calculated for >= 10 adverse events
ROR with 95% CI for each event
IC with 95% CI for each event
Signal strength classified for each
Demographics stratified for strong signals

已为>=10个不良事件计算PRR
已为每个事件计算带95%置信区间的ROR
已为每个事件计算带95%置信区间的IC
已为每个事件分类信号强度
已为强信号进行人口统计学分层

Phase 3: FDA Label

Phase 3: FDA说明书

Boxed warnings checked (or confirmed none)
Contraindications extracted
Warnings and precautions extracted
Adverse reactions from label
Drug interactions from label
Special populations (pregnancy, geriatric, pediatric)

已检查黑框警告（或确认无）
已提取禁忌症
已提取警告与注意事项
已提取说明书中的不良反应
已提取说明书中的药物相互作用
已提取特殊人群信息（妊娠、老年、儿科）

Phase 4: Mechanism Context

Phase 4: 作用机制关联

Target safety profile (OpenTargets)
OpenTargets adverse events queried
ADMET predictions (if SMILES available)

已获取靶点安全特征（OpenTargets）
已查询OpenTargets不良事件
已进行ADMET预测（若SMILES可用）

Phase 5: Comparative Analysis

Phase 5: 对比分析

At least 1 class comparison performed
Class-wide vs drug-specific signals identified
Aggregate class AEs computed (if applicable)

已完成至少1项同类药物对比
已识别同类共性信号与药物特异性信号
已计算同类药物汇总不良事件（若适用）

Phase 6: DDIs & PGx

Phase 6: DDI与PGx

DDIs from FDA label extracted
PharmGKB queried
Dosing guidelines checked
FDA PGx biomarkers checked

已提取FDA说明书中的DDI
已查询PharmGKB
已检查给药指南
已检查FDA PGx生物标志物

Phase 7: Literature

Phase 7: 文献

PubMed searched (>=10 articles)
OpenAlex citation analysis (if time permits)
Key safety publications cited

已搜索PubMed（>=10篇文献）
已进行OpenAlex引用分析（若时间允许）
已引用关键安全出版物

Phase 8: Risk Assessment

Phase 8: 风险评估

Safety Signal Score calculated (0-100)
Each signal evidence-graded (T1-T4)
Score interpretation provided

已计算安全信号评分（0-100分）
已为每个信号进行证据分级（T1-T4）
已提供评分解读

Phase 9: Report

Phase 9: 报告

Report file created and saved
Executive summary written
Monitoring recommendations provided
Risk mitigation strategies listed
Patient counseling points included
All sources cited

Tool Parameter Reference (Verified)

工具参数参考（已验证）

FAERS Tools (OpenFDA-based)

FAERS工具（基于OpenFDA）

Tool	Key Parameters	Notes
`FAERS_count_reactions_by_drug_event`	`medicinalproduct` (REQUIRED), `patientsex` , `patientagegroup` , `occurcountry`	Returns [{term, count}]
`FAERS_count_seriousness_by_drug_event`	`medicinalproduct` (REQUIRED), `patientsex` , `patientagegroup` , `occurcountry`	Returns [{term: "Serious"/"Non-serious", count}]
`FAERS_count_outcomes_by_drug_event`	`medicinalproduct` (REQUIRED), `patientsex` , `patientagegroup` , `occurcountry`	Returns [{term: "Fatal"/"Recovered"/..., count}]
`FAERS_count_patient_age_distribution`	`medicinalproduct` (REQUIRED)	Returns [{term: "Elderly"/"Adult"/..., count}]
`FAERS_count_death_related_by_drug`	`medicinalproduct` (REQUIRED)	Returns [{term: "alive"/"death", count}]
`FAERS_count_reportercountry_by_drug_event`	`medicinalproduct` (REQUIRED), `patientsex` , `patientagegroup` , `serious`	Returns [{term: "US"/"GB"/..., count}]
`FAERS_search_adverse_event_reports`	`medicinalproduct` , `limit` (max 100), `skip`	Returns individual case reports with patient/drug/reaction data
`FAERS_search_reports_by_drug_and_reaction`	`medicinalproduct` (REQUIRED), `reactionmeddrapt` (REQUIRED), `limit` , `skip` , `patientsex` , `serious`	Returns individual reports filtered by specific reaction
`FAERS_search_serious_reports_by_drug`	`medicinalproduct` (REQUIRED), `seriousnessdeath` , `seriousnesshospitalization` , `seriousnesslifethreatening` , `seriousnessdisabling` , `limit`	Returns serious event reports

工具	关键参数	说明
`FAERS_count_reactions_by_drug_event`	`medicinalproduct` （必填）, `patientsex` , `patientagegroup` , `occurcountry`	返回[{term, count}]
`FAERS_count_seriousness_by_drug_event`	`medicinalproduct` （必填）, `patientsex` , `patientagegroup` , `occurcountry`	返回[{term: "Serious"/"Non-serious", count}]
`FAERS_count_outcomes_by_drug_event`	`medicinalproduct` （必填）, `patientsex` , `patientagegroup` , `occurcountry`	返回[{term: "Fatal"/"Recovered"/..., count}]
`FAERS_count_patient_age_distribution`	`medicinalproduct` （必填）	返回[{term: "Elderly"/"Adult"/..., count}]
`FAERS_count_death_related_by_drug`	`medicinalproduct` （必填）	返回[{term: "alive"/"death", count}]
`FAERS_count_reportercountry_by_drug_event`	`medicinalproduct` （必填）, `patientsex` , `patientagegroup` , `serious`	返回[{term: "US"/"GB"/..., count}]
`FAERS_search_adverse_event_reports`	`medicinalproduct` , `limit` （最大100）, `skip`	返回包含患者/药物/反应数据的单个病例报告
`FAERS_search_reports_by_drug_and_reaction`	`medicinalproduct` （必填）, `reactionmeddrapt` （必填）, `limit` , `skip` , `patientsex` , `serious`	返回按特定反应筛选的单个报告
`FAERS_search_serious_reports_by_drug`	`medicinalproduct` （必填）, `seriousnessdeath` , `seriousnesshospitalization` , `seriousnesslifethreatening` , `seriousnessdisabling` , `limit`	返回严重事件报告

FAERS Analytics Tools (operation-based)

FAERS分析工具（基于operation）

Tool	Key Parameters	Notes
`FAERS_calculate_disproportionality`	`operation` ="calculate_disproportionality", `drug_name` (REQUIRED), `adverse_event` (REQUIRED)	Returns PRR, ROR, IC with 95% CI and signal detection
`FAERS_analyze_temporal_trends`	`operation` ="analyze_temporal_trends", `drug_name` (REQUIRED), `adverse_event` (optional)	Returns yearly counts and trend direction
`FAERS_compare_drugs`	`operation` ="compare_drugs", `drug1` (REQUIRED), `drug2` (REQUIRED), `adverse_event` (REQUIRED)	Returns PRR/ROR/IC for both drugs side-by-side
`FAERS_filter_serious_events`	`operation` ="filter_serious_events", `drug_name` (REQUIRED), `seriousness_type` (death/hospitalization/disability/life_threatening/all)	Returns top serious reactions with counts
`FAERS_stratify_by_demographics`	`operation` ="stratify_by_demographics", `drug_name` (REQUIRED), `adverse_event` (REQUIRED), `stratify_by` (sex/age/country)	Returns stratified counts and percentages. Sex codes: 0=Unknown, 1=Male, 2=Female
`FAERS_rollup_meddra_hierarchy`	`operation` ="rollup_meddra_hierarchy", `drug_name` (REQUIRED)	Returns top 50 preferred terms with counts

工具	关键参数	说明
`FAERS_calculate_disproportionality`	`operation` ="calculate_disproportionality", `drug_name` （必填）, `adverse_event` （必填）	返回PRR、ROR、IC及95%置信区间和信号检测结果
`FAERS_analyze_temporal_trends`	`operation` ="analyze_temporal_trends", `drug_name` （必填）, `adverse_event` （可选）	返回年度报告数及趋势方向
`FAERS_compare_drugs`	`operation` ="compare_drugs", `drug1` （必填）, `drug2` （必填）, `adverse_event` （必填）	返回两种药物的PRR/ROR/IC对比结果
`FAERS_filter_serious_events`	`operation` ="filter_serious_events", `drug_name` （必填）, `seriousness_type` （death/hospitalization/disability/life_threatening/all）	返回top严重反应及报告数
`FAERS_stratify_by_demographics`	`operation` ="stratify_by_demographics", `drug_name` （必填）, `adverse_event` （必填）, `stratify_by` （sex/age/country）	返回分层报告数及占比。性别编码: 0=未知, 1=男性, 2=女性
`FAERS_rollup_meddra_hierarchy`	`operation` ="rollup_meddra_hierarchy", `drug_name` （必填）	返回top 50首选术语及报告数

FAERS Aggregate Tools (multi-drug)

FAERS汇总工具（多药物）

Tool Key Parameters Notes

FAERS_count_additive_adverse_reactions

medicinalproducts

(REQUIRED, array),

patientsex

,

patientagegroup

,

occurcountry

,

serious

,

seriousnessdeath

Aggregates AE counts across multiple drugs

FAERS_count_additive_seriousness_classification

medicinalproducts

(REQUIRED, array),

patientsex

,

patientagegroup

,

occurcountry

Aggregates seriousness across multiple drugs

FAERS_count_additive_reaction_outcomes

medicinalproducts

(REQUIRED, array)

Aggregates outcomes across multiple drugs

工具关键参数说明

FAERS_count_additive_adverse_reactions

medicinalproducts

（必填，数组）,

patientsex

,

patientagegroup

,

occurcountry

,

serious

,

seriousnessdeath

汇总多个药物的不良事件报告数

FAERS_count_additive_seriousness_classification

medicinalproducts

（必填，数组）,

patientsex

,

patientagegroup

,

occurcountry

汇总多个药物的严重程度分布

FAERS_count_additive_reaction_outcomes

medicinalproducts

（必填，数组）

汇总多个药物的结局分布

FDA Label Tools

FDA说明书工具

Tool	Key Parameters	Notes
`FDA_get_boxed_warning_info_by_drug_name`	`drug_name`	Returns `{error: {code: "NOT_FOUND"}}` if no boxed warning
`FDA_get_contraindications_by_drug_name`	`drug_name`	Returns `{meta: {total: N}, results: [{contraindications: [...]}]}`
`FDA_get_adverse_reactions_by_drug_name`	`drug_name`	Returns `{meta: {total: N}, results: [{adverse_reactions: [...]}]}`
`FDA_get_warnings_by_drug_name`	`drug_name`	Returns `{meta: {total: N}, results: [{warnings: [...]}]}`
`FDA_get_drug_interactions_by_drug_name`	`drug_name`	Returns `{meta: {total: N}, results: [{drug_interactions: [...]}]}`
`FDA_get_pharmacogenomics_info_by_drug_name`	`drug_name`	Returns PGx info from label
`FDA_get_pregnancy_or_breastfeeding_info_by_drug_name`	`drug_name`	Returns pregnancy info
`FDA_get_geriatric_use_info_by_drug_name`	`drug_name`	Returns geriatric use info
`FDA_get_pediatric_use_info_by_drug_name`	`drug_name`	Returns pediatric info

工具	关键参数	说明
`FDA_get_boxed_warning_info_by_drug_name`	`drug_name`	若无黑框警告，返回 `{error: {code: "NOT_FOUND"}}`
`FDA_get_contraindications_by_drug_name`	`drug_name`	返回 `{meta: {total: N}, results: [{contraindications: [...]}]}`
`FDA_get_adverse_reactions_by_drug_name`	`drug_name`	返回 `{meta: {total: N}, results: [{adverse_reactions: [...]}]}`
`FDA_get_warnings_by_drug_name`	`drug_name`	返回 `{meta: {total: N}, results: [{warnings: [...]}]}`
`FDA_get_drug_interactions_by_drug_name`	`drug_name`	返回 `{meta: {total: N}, results: [{drug_interactions: [...]}]}`
`FDA_get_pharmacogenomics_info_by_drug_name`	`drug_name`	返回说明书中的PGx信息
`FDA_get_pregnancy_or_breastfeeding_info_by_drug_name`	`drug_name`	返回妊娠相关信息
`FDA_get_geriatric_use_info_by_drug_name`	`drug_name`	返回老年用药信息
`FDA_get_pediatric_use_info_by_drug_name`	`drug_name`	返回儿科用药信息

OpenTargets Tools

OpenTargets工具

Tool	Key Parameters	Notes
`OpenTargets_get_drug_chembId_by_generic_name`	`drugName`	Returns `{data: {search: {hits: [{id, name, description}]}}}`
`OpenTargets_get_drug_adverse_events_by_chemblId`	`chemblId`	Returns `{data: {drug: {adverseEvents: {count, rows: [{name, meddraCode, count, logLR}]}}}}`
`OpenTargets_get_drug_blackbox_status_by_chembl_ID`	`chemblId`	Returns `{data: {drug: {hasBeenWithdrawn, blackBoxWarning}}}`
`OpenTargets_get_drug_warnings_by_chemblId`	`chemblId`	Returns drug warnings (may be empty)
`OpenTargets_get_drug_mechanisms_of_action_by_chemblId`	`chemblId`	Returns `{data: {drug: {mechanismsOfAction: {rows: [{mechanismOfAction, actionType, targetName, targets}]}}}}`
`OpenTargets_get_drug_indications_by_chemblId`	`chemblId`	Returns approved and investigational indications
`OpenTargets_get_target_safety_profile_by_ensemblID`	`ensemblId`	Returns `{data: {target: {safetyLiabilities: [{event, effects, studies, datasource}]}}}`

工具	关键参数	说明
`OpenTargets_get_drug_chembId_by_generic_name`	`drugName`	返回 `{data: {search: {hits: [{id, name, description}]}}}`
`OpenTargets_get_drug_adverse_events_by_chemblId`	`chemblId`	返回 `{data: {drug: {adverseEvents: {count, rows: [{name, meddraCode, count, logLR}]}}}}`
`OpenTargets_get_drug_blackbox_status_by_chembl_ID`	`chemblId`	返回 `{data: {drug: {hasBeenWithdrawn, blackBoxWarning}}}`
`OpenTargets_get_drug_warnings_by_chemblId`	`chemblId`	返回药物警告（可能为空）
`OpenTargets_get_drug_mechanisms_of_action_by_chemblId`	`chemblId`	返回 `{data: {drug: {mechanismsOfAction: {rows: [{mechanismOfAction, actionType, targetName, targets}]}}}}`
`OpenTargets_get_drug_indications_by_chemblId`	`chemblId`	返回获批及在研适应症
`OpenTargets_get_target_safety_profile_by_ensemblID`	`ensemblId`	返回 `{data: {target: {safetyLiabilities: [{event, effects, studies, datasource}]}}}`

DrugBank Tools

DrugBank工具

Tool	Key Parameters	Notes
`drugbank_get_safety_by_drug_name_or_drugbank_id`	`query` , `case_sensitive` (bool), `exact_match` (bool), `limit`	Returns toxicity, food interactions
`drugbank_get_targets_by_drug_name_or_drugbank_id`	`query` , `case_sensitive` , `exact_match` , `limit`	Returns drug targets
`drugbank_get_drug_interactions_by_drug_name_or_id`	`query` , `case_sensitive` , `exact_match` , `limit`	Returns DDIs
`drugbank_get_pharmacology_by_drug_name_or_drugbank_id`	`query` , `case_sensitive` , `exact_match` , `limit`	Returns pharmacology

工具	关键参数	说明
`drugbank_get_safety_by_drug_name_or_drugbank_id`	`query` , `case_sensitive` （布尔值）, `exact_match` （布尔值）, `limit`	返回毒性、食物相互作用信息
`drugbank_get_targets_by_drug_name_or_drugbank_id`	`query` , `case_sensitive` , `exact_match` , `limit`	返回药物靶点
`drugbank_get_drug_interactions_by_drug_name_or_id`	`query` , `case_sensitive` , `exact_match` , `limit`	返回药物相互作用
`drugbank_get_pharmacology_by_drug_name_or_drugbank_id`	`query` , `case_sensitive` , `exact_match` , `limit`	返回药理学信息

PharmGKB Tools

PharmGKB工具

Tool	Key Parameters	Notes
`PharmGKB_search_drugs`	`query`	Returns `{data: [{id, name, smiles}]}`
`PharmGKB_get_drug_details`	`drug_id` (e.g., "PA448500")	Returns detailed drug info
`PharmGKB_get_dosing_guidelines`	`guideline_id` , `gene` (both optional)	Returns dosing guidelines
`PharmGKB_get_clinical_annotations`	`annotation_id` , `gene_id` (both optional)	Returns clinical annotations
`fda_pharmacogenomic_biomarkers`	`drug_name` , `biomarker` , `limit`	Returns `{count, results: [...]}`

工具	关键参数	说明
`PharmGKB_search_drugs`	`query`	返回 `{data: [{id, name, smiles}]}`
`PharmGKB_get_drug_details`	`drug_id` （例如"PA448500"）	返回详细药物信息
`PharmGKB_get_dosing_guidelines`	`guideline_id` , `gene` （均为可选）	返回给药指南
`PharmGKB_get_clinical_annotations`	`annotation_id` , `gene_id` （均为可选）	返回临床注释
`fda_pharmacogenomic_biomarkers`	`drug_name` , `biomarker` , `limit`	返回 `{count, results: [...]}`

ADMETAI Tools

ADMETAI工具

Tool	Key Parameters	Notes
`ADMETAI_predict_toxicity`	`smiles` (REQUIRED, array of strings)	Predicts hepatotoxicity, cardiotoxicity, etc.
`ADMETAI_predict_CYP_interactions`	`smiles` (REQUIRED, array)	Predicts CYP inhibition/substrate

工具	关键参数	说明
`ADMETAI_predict_toxicity`	`smiles` （必填，字符串数组）	预测肝毒性、心脏毒性等
`ADMETAI_predict_CYP_interactions`	`smiles` （必填，数组）	预测CYP抑制/底物

Literature Tools

文献工具

Tool	Key Parameters	Notes
`PubMed_search_articles`	`query` , `limit`	Returns list of article dicts
`openalex_search_works`	`query` , `limit`	Returns works with citation counts
`EuropePMC_search_articles`	`query` , `source` ("PPR" for preprints), `pageSize`	Returns articles including preprints
`search_clinical_trials`	`query_term` (REQUIRED), `condition` , `intervention` , `pageSize`	Returns clinical trials

工具	关键参数	说明
`PubMed_search_articles`	`query` , `limit`	返回文章字典列表
`openalex_search_works`	`query` , `limit`	返回带有引用数的文献
`EuropePMC_search_articles`	`query` , `source` （"PPR"表示预印本）, `pageSize`	返回包含预印本的文章
`search_clinical_trials`	`query_term` （必填）, `condition` , `intervention` , `pageSize`	返回临床试验信息

Fallback Chains

备选工具链

Primary Tool	Fallback 1	Fallback 2
`FAERS_calculate_disproportionality`	Manual calculation from `FAERS_count_*` data	Literature PRR values
`FAERS_count_reactions_by_drug_event`	`FAERS_rollup_meddra_hierarchy`	OpenTargets adverse events
`FDA_get_boxed_warning_info_by_drug_name`	`OpenTargets_get_drug_blackbox_status_by_chembl_ID`	DrugBank safety
`FDA_get_contraindications_by_drug_name`	`FDA_get_warnings_by_drug_name`	DrugBank safety
`OpenTargets_get_drug_chembId_by_generic_name`	`ChEMBL_search_drugs`	Manual search
`PharmGKB_search_drugs`	`fda_pharmacogenomic_biomarkers`	FDA label PGx section
`PubMed_search_articles`	`openalex_search_works`	`EuropePMC_search_articles`

主工具	备选工具1	备选工具2
`FAERS_calculate_disproportionality`	基于 `FAERS_count_*` 数据手动计算	文献中的PRR值
`FAERS_count_reactions_by_drug_event`	`FAERS_rollup_meddra_hierarchy`	OpenTargets不良事件
`FDA_get_boxed_warning_info_by_drug_name`	`OpenTargets_get_drug_blackbox_status_by_chembl_ID`	DrugBank安全信息
`FDA_get_contraindications_by_drug_name`	`FDA_get_warnings_by_drug_name`	DrugBank安全信息
`OpenTargets_get_drug_chembId_by_generic_name`	`ChEMBL_search_drugs`	手动搜索
`PharmGKB_search_drugs`	`fda_pharmacogenomic_biomarkers`	FDA说明书PGx章节
`PubMed_search_articles`	`openalex_search_works`	`EuropePMC_search_articles`

Common Patterns

常见模式

Pattern 1: Full Safety Signal Profile for a Single Drug

模式1: 单药完整安全信号分析

Use all phases (0-9) for comprehensive report. Best for regulatory submissions, safety reviews.

使用所有Phase（0-9）生成全面报告。适用于监管申报、安全性评审。

Pattern 2: Specific Adverse Event Investigation

模式2: 特定不良事件调查

Focus on Phases 0, 2, 3, 7. User asks "Does [drug] cause [event]?" - calculate disproportionality for that specific event, check label, search literature.

重点关注Phase 0、2、3、7。当用户询问“[药物]是否会导致[事件]？”时——计算该特定事件的比例失衡指标，检查说明书，搜索文献。

Pattern 3: Drug Class Comparison

模式3: 同类药物对比

Focus on Phases 0, 2, 5. Compare 3-5 drugs in same class for a specific adverse event using

FAERS_compare_drugs

.

重点关注Phase 0、2、5。使用

FAERS_compare_drugs

对比3-5个同类药物的特定不良事件。

Pattern 4: Emerging Signal Detection

模式4: 新出现信号检测

Focus on Phases 1, 2, 7. Screen top 20+ FAERS events for signals, identify any not in FDA label (Phase 3), search recent literature for confirmation.

重点关注Phase 1、2、7。筛选FAERS top 20+事件以识别信号，找出FDA说明书中未提及的事件（Phase 3），搜索近期文献确认。

Pattern 5: Pharmacogenomic Risk Assessment

模式5: 药物基因组学风险评估

Focus on Phases 0, 6. Identify genetic risk factors for adverse events using PharmGKB and FDA PGx biomarkers.

重点关注Phase 0、6。利用PharmGKB和FDA PGx生物标志物识别不良事件的遗传风险因素。

Pattern 6: Pre-Approval Safety Assessment

模式6: 获批前安全性评估

Focus on Phases 4, 7. Use ADMET predictions and target safety profiles when FAERS data is limited (new drugs).

重点关注Phase 4、7。当FAERS数据有限时（新药），使用ADMET预测和靶点安全特征。

Edge Cases

边缘情况

Drug with No FAERS Reports

无FAERS报告的药物

Skip Phases 1-2
Rely on FDA label (Phase 3), mechanism predictions (Phase 4), and literature (Phase 7)
Safety Signal Score will be lower due to lack of signal detection data

跳过Phase 1-2
依赖FDA说明书（Phase 3）、机制预测（Phase 4）和文献（Phase 7）
由于缺乏信号检测数据，安全信号评分会较低

Generic vs Brand Name

通用名 vs 商品名

Always try both names in FAERS queries (FAERS uses brand names sometimes)

Use

OpenTargets_get_drug_chembId_by_generic_name

to resolve to standard identifier

Use
```
FDA_get_brand_name_generic_name
```
for name cross-reference

FAERS查询时始终尝试两种名称（FAERS有时使用商品名）

使用

OpenTargets_get_drug_chembId_by_generic_name

关联至标准标识符

使用
```
FDA_get_brand_name_generic_name
```
进行名称交叉验证

Drug Combinations

联合用药

Use

FAERS_search_reports_by_drug_combination

for polypharmacy analysis

Distinguish combination AEs from individual drug AEs
Use
```
FAERS_count_additive_adverse_reactions
```
for aggregate class analysis

使用

FAERS_search_reports_by_drug_combination

进行多药分析

区分联合用药不良事件与单药不良事件
使用
```
FAERS_count_additive_adverse_reactions
```
进行同类药物汇总分析

Confounding by Indication

适应症混杂

Compare AE profile to the disease being treated
Example: "Death" reports for chemotherapy drugs may reflect disease progression
Always note this limitation in the report

将不良事件谱与治疗疾病对比
示例：化疗药物的“死亡”报告可能反映疾病进展
报告中需始终注明该局限性

Drugs with Boxed Warnings

带有黑框警告的药物

Score component automatically 25/25 for label warnings
Prioritize boxed warning events in disproportionality analysis
Cross-reference boxed warning with FAERS signal strength

说明书警告成分自动得25/25分
在比例失衡分析中优先关注黑框警告事件
交叉验证黑框警告与FAERS信号强度

tooluniverse-adverse-event-detection

Original

Translation

Adverse Drug Event Signal Detection & Analysis

药物不良事件信号检测与分析

When to Use

适用场景

Workflow Overview

工作流程概述

Phase 0: Input Parsing & Drug Disambiguation

Phase 0: 输入解析与药物消歧

0.1 Resolve Drug Identity

0.1 确认药物身份

Step 1: Get ChEMBL ID from drug name

Step 1: 通过药名获取ChEMBL ID

Response: {data: {search: {hits: [{id: "CHEMBL1487", name: "ATORVASTATIN", description: "..."}]}}}

Response: {data: {search: {hits: [{id: "CHEMBL1487", name: "ATORVASTATIN", description: "..."}]}}}

Step 2: Get drug mechanism of action

Step 2: 获取药物作用机制

Response: {data: {drug: {mechanismsOfAction: {rows: [{mechanismOfAction: "HMG-CoA reductase inhibitor", actionType: "INHIBITOR", targetName: "...", targets: [{id: "ENSG00000113161", approvedSymbol: "HMGCR"}]}]}}}}

Response: {data: {drug: {mechanismsOfAction: {rows: [{mechanismOfAction: "HMG-CoA reductase inhibitor", actionType: "INHIBITOR", targetName: "...", targets: [{id: "ENSG00000113161", approvedSymbol: "HMGCR"}]}]}}}}

Step 3: Get blackbox warning status

Step 3: 获取黑框警告状态

Response: {data: {drug: {name: "ATORVASTATIN", hasBeenWithdrawn: false, blackBoxWarning: false}}}

Response: {data: {drug: {name: "ATORVASTATIN", hasBeenWithdrawn: false, blackBoxWarning: false}}}

Step 4: Get DrugBank info (safety, toxicity)

Step 4: 获取DrugBank信息（安全性、毒性）

Response: {results: [{drug_name: "Atorvastatin", drugbank_id: "DB01076", toxicity: "...", food_interactions: "..."}]}

Response: {results: [{drug_name: "Atorvastatin", drugbank_id: "DB01076", toxicity: "...", food_interactions: "..."}]}

Step 5: Get DrugBank targets

Step 5: 获取DrugBank靶点

Response: {results: [{drug_name: "...", targets: [{name: "HMG-CoA reductase", ...}]}]}

Response: {results: [{drug_name: "...", targets: [{name: "HMG-CoA reductase", ...}]}]}

Step 6: Get approved indications

Step 6: 获取获批适应症

Response: {data: {drug: {indications: {rows: [{disease: {name: "hypercholesterolemia"}, maxPhaseForIndication: 4}]}}}}

Response: {data: {drug: {indications: {rows: [{disease: {name: "hypercholesterolemia"}, maxPhaseForIndication: 4}]}}}}

0.2 Output for Report

0.2 报告输出内容

1. Drug Identification

1. 药物识别

Phase 1: FAERS Adverse Event Profiling

Phase 1: FAERS不良事件分析

1.1 Query FAERS for Adverse Events

1.1 查询FAERS不良事件

Get top adverse event reactions (returns list of {term, count})

获取top不良事件反应（返回{term, count}列表）

Response: [{term: "FATIGUE", count: 19171}, {term: "DIARRHOEA", count: 17127}, ...]

Response: [{term: "FATIGUE", count: 19171}, {term: "DIARRHOEA", count: 17127}, ...]

Get seriousness classification

获取严重程度分类

Response: [{term: "Serious", count: 242757}, {term: "Non-serious", count: 83504}]

Response: [{term: "Serious", count: 242757}, {term: "Non-serious", count: 83504}]

Get outcome distribution

获取结局分布

Response: [{term: "Unknown", count: 162310}, {term: "Fatal", count: 22128}, ...]

Response: [{term: "Unknown", count: 162310}, {term: "Fatal", count: 22128}, ...]

Get age distribution

获取年龄分布

Response: [{term: "Elderly", count: 38510}, {term: "Adult", count: 24302}, ...]

Response: [{term: "Elderly", count: 38510}, {term: "Adult", count: 24302}, ...]

Get death-related events

获取死亡相关事件

Response: [{term: "alive", count: 113157}, {term: "death", count: 26909}]

Response: [{term: "alive", count: 113157}, {term: "death", count: 26909}]

Get reporter country distribution

获取报告者国家分布

Response: [{term: "US", count: 170963}, {term: "GB", count: 40079}, ...]

Response: [{term: "US", count: 170963}, {term: "GB", count: 40079}, ...]

1.2 Get Serious Events Breakdown

1.2 严重事件细分

Filter serious events - all types

筛选所有类型的严重事件

Response: {status: "success", drug_name: "ATORVASTATIN", seriousness_type: "all",

Response: {status: "success", drug_name: "ATORVASTATIN", seriousness_type: "all",

total_serious_events: N, top_serious_reactions: [{reaction: "...", count: N}, ...]}

total_serious_events: N, top_serious_reactions: [{reaction: "...", count: N}, ...]}

Death-related serious events

死亡相关严重事件

Response: {status: "success", total_serious_events: 18720,