tooluniverse-clinical-data-integration

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Clinical Data Integration for Drug Safety

面向药物安全的临床数据整合

End-to-end drug safety review pipeline that integrates FDA label information, FAERS spontaneous reports, disproportionality signal detection, pharmacogenomic biomarkers, clinical trial data, and published literature. Designed for regulatory assessments, pharmacovigilance, and clinical decision support.

Guiding principles:

Label is ground truth -- FDA-approved labeling is the authoritative starting point for known safety information
Signals need context -- a FAERS signal without label or literature corroboration is hypothesis-generating, not confirmatory
Disproportionality is not causation -- PRR/ROR measure reporting patterns, not causal relationships
Pharmacogenomics narrows risk -- PGx biomarkers can identify which patients face elevated risk
Progressive reporting -- create the report file early; update section by section
English-first queries -- use English drug names in all tool calls; respond in the user's language

Clinical data integration starts with data harmonization. Different hospitals code the same diagnosis differently (ICD-10 vs SNOMED). Before merging datasets, verify the coding system. Missing data is informative — a missing lab value may mean the test wasn't ordered (patient was stable) not that the result was normal.

端到端的药物安全审查流程，整合了FDA标签信息、FAERS自发报告、不成比例性信号检测、药物基因组学生物标志物、临床试验数据及已发表文献。专为监管评估、药物警戒及临床决策支持设计。

指导原则:

标签为基准事实——FDA批准的标签是已知安全信息的权威起点
信号需结合上下文——缺乏标签或文献佐证的FAERS信号仅用于生成假设，而非确认结论
不成比例性不等于因果关系——PRR/ROR衡量的是报告模式，而非因果关系
药物基因组学缩小风险范围——PGx生物标志物可识别面临高风险的患者群体
渐进式报告——尽早创建报告文件，逐节更新内容
优先使用英文查询——所有工具调用中使用英文药物名称；以用户使用的语言回复

临床数据整合始于数据协调。不同医院对同一诊断的编码方式不同（ICD-10 vs SNOMED）。合并数据集前，请验证编码系统。缺失数据具有参考价值——缺失的实验室值可能意味着未开具相关检测（患者状态稳定），而非结果正常。

LOOK UP, DON'T GUESS

查资料，勿猜测

When uncertain about any scientific fact, SEARCH databases first rather than reasoning from memory. A database-verified answer is always more reliable than a guess.

Differentiation: This skill emphasizes regulatory-grade data integration across the full drug lifecycle. For focused FAERS signal detection with quantitative scoring, see

tooluniverse-adverse-event-detection

. For general pharmacovigilance workflows, see

tooluniverse-pharmacovigilance

当对任何科学事实不确定时，先搜索数据库，而非凭记忆推理。经数据库验证的答案永远比猜测更可靠。

差异化说明：本技能强调覆盖药物全生命周期的监管级数据整合。如需聚焦FAERS信号检测并进行定量评分，请查看

tooluniverse-adverse-event-detection

。如需通用药物警戒工作流程，请查看

tooluniverse-pharmacovigilance

。

COMPUTE, DON'T DESCRIBE

计算，勿描述

When analysis requires computation (statistics, data processing, scoring, enrichment), write and run Python code via Bash. Don't describe what you would do — execute it and report actual results. Use ToolUniverse tools to retrieve data, then Python (pandas, scipy, statsmodels, matplotlib) to analyze it.

当分析需要计算（统计、数据处理、评分、富集分析）时，通过Bash编写并运行Python代码。不要描述你会做什么——直接执行并报告实际结果。使用ToolUniverse工具检索数据，然后用Python（pandas、scipy、statsmodels、matplotlib）进行分析。

When to Use

使用场景

Typical triggers:

"Give me a full safety review for [drug]"
"What does the FDA label say about [drug] and [event]?"
"Are there FAERS signals for [drug]?"
"What pharmacogenomic biomarkers exist for [drug]?"
"Find clinical trials studying [drug] safety"
"Post-market surveillance summary for [drug]"
"Compare safety profiles of [drug A] and [drug B]"

典型触发场景:

"为[药物]提供完整的安全审查"
"FDA标签中关于[药物]和[事件]的内容是什么？"
"[药物]是否存在FAERS信号？"
"[药物]有哪些药物基因组学生物标志物？"
"查找研究[药物]安全性的临床试验"
"[药物]的上市后监测总结"
"比较[药物A]和[药物B]的安全性特征"

Core Data Sources

核心数据源

Source	Type	Best For
FDA Labels (DailyMed)	Regulatory	Approved safety information, boxed warnings, drug interactions
FAERS	Spontaneous reports	Post-market adverse event signals, demographic patterns
CPIC	Guidelines	Pharmacogenomic dosing recommendations
FDA PGx Biomarkers	Regulatory	Approved pharmacogenomic labeling
ClinicalTrials.gov	Trial registry	Ongoing/completed safety trials
PubMed	Literature	Published safety studies, case reports

来源	类型	最佳用途
FDA Labels (DailyMed)	监管类	已批准的安全信息、黑框警告、药物相互作用
FAERS	自发报告类	上市后不良事件信号、人口统计学模式
CPIC	指南类	药物基因组学给药建议
FDA PGx Biomarkers	监管类	已批准的药物基因组学标签
ClinicalTrials.gov	试验注册类	进行中/已完成的安全性试验
PubMed	文献类	已发表的安全性研究、病例报告

Workflow Overview

工作流程概述

Phase 0: Drug Identity & Context
  Resolve drug name, get class, mechanism, indications
    |
Phase 1: FDA Label Extraction
  Boxed warnings, contraindications, adverse reactions, interactions
    |
Phase 2: FAERS Signal Detection
  Top adverse events, disproportionality (PRR/ROR), demographics
    |
Phase 3: Pharmacogenomics
  CPIC guidelines, FDA PGx biomarkers, genotype-specific risks
    |
Phase 4: Clinical Trials
  Safety-focused trials, risk evaluation programs
    |
Phase 5: Literature Evidence
  PubMed safety studies, case reports, meta-analyses
    |
Phase 6: Integrated Safety Report
  Synthesize all sources into a cohesive safety profile

Phase 0: Drug Identity & Context
  Resolve drug name, get class, mechanism, indications
    |
Phase 1: FDA Label Extraction
  Boxed warnings, contraindications, adverse reactions, interactions
    |
Phase 2: FAERS Signal Detection
  Top adverse events, disproportionality (PRR/ROR), demographics
    |
Phase 3: Pharmacogenomics
  CPIC guidelines, FDA PGx biomarkers, genotype-specific risks
    |
Phase 4: Clinical Trials
  Safety-focused trials, risk evaluation programs
    |
Phase 5: Literature Evidence
  PubMed safety studies, case reports, meta-analyses
    |
Phase 6: Integrated Safety Report
  Synthesize all sources into a cohesive safety profile

Phase Details

阶段详情

Phase 0: Drug Identity & Context

Phase 0: 药物识别与背景信息

Objective: Unambiguously identify the drug and establish baseline context.

Tools:

```
DailyMed_search_spls
```
-- search Structured Product Labels
- Input:
```
query
```
  (drug name)
- Output: SPL list with set IDs, titles, labeler names
```
OpenFDA_get_approval_history
```
-- get approval dates and supplements
- Input:
```
drug_name
```
  (generic or brand name)
- Output: approval dates, application numbers, supplement history

Workflow:

Search DailyMed to confirm the drug name and identify the correct SPL
Get approval history to establish how long the drug has been marketed
Note the therapeutic class, mechanism of action, and approved indications
Record brand names vs generic name for consistent FAERS queries

Tip: FAERS uses

medicinalproduct

which can be brand or generic. Try both forms in Phase 2.

Objective: 明确识别药物并建立基线背景信息。

Tools:

```
DailyMed_search_spls
```
——搜索结构化产品标签（SPL）
- Input:
```
query
```
  （药物名称）
- Output: 包含set ID、标题、标签方名称的SPL列表
```
OpenFDA_get_approval_history
```
——获取批准日期及补充信息
- Input:
```
drug_name
```
  （通用名或商品名）
- Output: 批准日期、申请编号、补充历史记录

工作流程:

搜索DailyMed确认药物名称并识别正确的SPL
获取批准历史以确定药物上市时长
记录治疗类别、作用机制及获批适应症
记录商品名与通用名，确保FAERS查询的一致性

提示: FAERS使用

medicinalproduct

字段，可填写商品名或通用名。在Phase 2中尝试两种形式。

Phase 1: FDA Label Extraction

Phase 1: FDA标签提取

Objective: Extract all safety-relevant sections from the FDA-approved label.

Tools:

```
FDA_get_boxed_warning_info_by_drug_name
```
-- boxed (black box) warnings
- Input:
```
drug_name
```
- Output: warning text, or
```
{error: {code: "NOT_FOUND"}}
```
  if none exists (normal)
```
FDA_get_warnings_and_cautions_by_drug_name
```
-- warnings and precautions section
- Input:
```
drug_name
```
- Output: full warnings text
```
DailyMed_parse_adverse_reactions
```
-- adverse reactions from label
- Input:
```
setid
```
  (NOT
```
set_id
```
  ; from Phase 0 DailyMed search)
- Output: parsed adverse reaction tables and text
```
DailyMed_parse_drug_interactions
```
-- drug interaction section
- Input:
```
setid
```
  (NOT
```
set_id
```
  )
- Output: parsed interaction data

Workflow:

Check for boxed warnings first -- these represent the most serious safety concerns
Extract warnings and precautions
Parse adverse reactions (both clinical trial rates and post-marketing reports)
Extract drug interactions
A
```
NOT_FOUND
```
response for boxed warnings is normal and means no boxed warning exists

Label section priority: Boxed Warning > Contraindications > Warnings/Precautions > Adverse Reactions > Drug Interactions

目标: 从FDA批准的标签中提取所有与安全相关的内容。

工具:

```
FDA_get_boxed_warning_info_by_drug_name
```
——黑框警告信息
- Input:
```
drug_name
```
- Output: 警告文本；若不存在则返回
```
{error: {code: "NOT_FOUND"}}
```
  （正常情况）
```
FDA_get_warnings_and_cautions_by_drug_name
```
——警告与注意事项部分
- Input:
```
drug_name
```
- Output: 完整警告文本
```
DailyMed_parse_adverse_reactions
```
——标签中的不良反应信息
- Input:
```
setid
```
  （注意：不是
```
set_id
```
  ；来自Phase 0的DailyMed搜索结果）
- Output: 解析后的不良反应表格及文本
```
DailyMed_parse_drug_interactions
```
——药物相互作用部分
- Input:
```
setid
```
  （注意：不是
```
set_id
```
  ）
- Output: 解析后的相互作用数据

工作流程:

首先检查黑框警告——这些代表最严重的安全问题
提取警告与注意事项
解析不良反应（包括临床试验发生率及上市后报告）
提取药物相互作用信息
黑框警告返回
```
NOT_FOUND
```
是正常情况，意味着不存在黑框警告

标签内容优先级: 黑框警告 > 禁忌症 > 警告/注意事项 > 不良反应 > 药物相互作用

Phase 2: FAERS Signal Detection

Phase 2: FAERS信号检测

Objective: Identify post-market safety signals from spontaneous reports.

Tools:

FAERS_count_reactions_by_drug_event

-- top adverse events by frequency

Input:
```
medicinalproduct
```
(drug name, NOT
```
drug_name
```
)
Output:
```
[{term, count}]
```

FAERS_calculate_disproportionality

-- PRR, ROR, IC for drug-event pair

Input:
```
drug_name
```
,
```
adverse_event
```

Output:

{metrics: {PRR: {value, ci_95_lower, ci_95_upper}, ROR: {...}, IC: {...}}, signal_detection: {signal_detected, signal_strength}}

```
FAERS_filter_serious_events
```
-- filter by seriousness type
- Input:
```
drug_name
```
  ,
```
seriousness_type
```
  (all/death/hospitalization/disability/life_threatening)
- Output: serious event breakdown
```
FAERS_stratify_by_demographics
```
-- age/sex/country stratification
- Input:
```
drug_name
```
  ,
```
adverse_event
```
  (optional),
```
stratify_by
```
  (sex/age/country)
- Output: demographic breakdown (sex codes: 0=Unknown, 1=Male, 2=Female)

Workflow:

Get top 20 adverse events by report count
For the top 10-15, calculate disproportionality (PRR, ROR, IC with 95% CI)
Signal criteria: PRR >= 2.0, lower CI > 1.0, N >= 3 reports
For detected signals, filter by seriousness (deaths, hospitalizations)
Stratify strong signals by demographics to identify at-risk populations

Important notes:

FAERS_count_reactions_by_drug_event

uses

medicinalproduct

param, not

drug_name

FAERS_calculate_disproportionality

uses

drug_name

param

MedDRA term levels differ between count and disproportionality tools; case counts may not match exactly

FAERS signal interpretation — what the numbers mean:

Metric	Value	Interpretation
PRR (Proportional Reporting Ratio)	< 1.0	Event reported LESS than expected (possible protective effect or underreporting)
	1.0-2.0	No signal or weak signal
	2.0-5.0	Moderate signal — warrants investigation
	> 5.0	Strong signal — likely real association (but still not proof of causation)
ROR (Reporting Odds Ratio)	Similar to PRR but accounts for all other drugs	Same thresholds as PRR; slightly more robust
IC (Information Component)	< 0	No signal
	0-2	Weak signal
	> 2	Strong signal

Signal ≠ Causation: A strong FAERS signal means the drug-event pair is reported more often than expected. This could be due to:

True causal relationship (most important)
Channeling bias (sicker patients get the drug)
Notoriety bias (media attention increases reporting)
Protopathic bias (drug prescribed for early symptoms of the event)

How to assess signal credibility:

Is the event in the FDA label? (Label confirmation = strongest evidence)
Is there a plausible mechanism? (Drug's pharmacology explains the event)
Is there a dose-response? (Higher doses → more events)
Is there temporal consistency? (Event occurs after drug start, resolves after stop)
Is there epidemiological confirmation? (Published case-control or cohort study)

目标: 从自发报告中识别上市后安全信号。

工具:

```
FAERS_count_reactions_by_drug_event
```
——按频率排序的主要不良事件
- Input:
```
medicinalproduct
```
  （药物名称，注意：不是
```
drug_name
```
  ）
- Output:
```
[{term, count}]
```

FAERS_calculate_disproportionality

——计算药物-事件对的PRR、ROR、IC值

Input:
```
drug_name
```
,
```
adverse_event
```

Output:

{metrics: {PRR: {value, ci_95_lower, ci_95_upper}, ROR: {...}, IC: {...}}, signal_detection: {signal_detected, signal_strength}}

```
FAERS_filter_serious_events
```
——按严重程度类型筛选
- Input:
```
drug_name
```
  ,
```
seriousness_type
```
  （all/death/hospitalization/disability/life_threatening）
- Output: 严重事件细分结果
```
FAERS_stratify_by_demographics
```
——按年龄/性别/国家分层分析
- Input:
```
drug_name
```
  ,
```
adverse_event
```
  （可选）,
```
stratify_by
```
  （sex/age/country）
- Output: 人口统计学细分结果（性别编码：0=未知，1=男性，2=女性）

工作流程:

获取报告数量排名前20的不良事件
对排名前10-15的事件计算不成比例性（PRR、ROR、IC及95%置信区间）
信号判定标准：PRR ≥ 2.0，95%置信区间下限 > 1.0，报告数量N ≥ 3
对检测到的信号，按严重程度筛选（死亡、住院）
对强信号进行人口统计学分层分析，识别高风险人群

重要说明:

FAERS_count_reactions_by_drug_event

使用

medicinalproduct

参数，而非

drug_name

FAERS_calculate_disproportionality

使用

drug_name

参数

计数工具与不成比例性工具使用的MedDRA术语级别不同，病例计数可能不完全匹配

FAERS信号解读——数值含义:

指标	数值	解读
PRR（比例报告比）	< 1.0	事件报告频率低于预期（可能存在保护作用或报告不足）
	1.0-2.0	无信号或弱信号
	2.0-5.0	中度信号——需开展调查
	> 5.0	强信号——可能存在真实关联（但仍无法证明因果关系）
ROR（报告比值比）	与PRR类似，但考虑了所有其他药物	阈值与PRR相同；鲁棒性略高
IC（信息成分）	< 0	无信号
	0-2	弱信号
	> 2	强信号

信号≠因果关系: 强FAERS信号意味着药物-事件对的报告频率高于预期。这可能由以下原因导致:

真实的因果关系（最重要）
分流偏倚（病情更重的患者使用该药物）
知名度偏倚（媒体关注度提高了报告率）
先发偏倚（药物用于治疗事件的早期症状）

如何评估信号可信度:

该事件是否在FDA标签中提及？（标签确认=最强证据）
是否存在合理的作用机制？（药物药理学可解释该事件）
是否存在剂量反应关系？（剂量越高→事件越多）
是否存在时间一致性？（事件发生在用药后，停药后缓解）
是否有流行病学证据支持？（已发表的病例对照或队列研究）

Phase 3: Pharmacogenomics

Phase 3: 药物基因组学

Objective: Identify genetic factors that modify drug safety.

Tools:

```
CPIC_list_guidelines
```
-- get CPIC pharmacogenomic guidelines
- Input: optional
```
gene
```
  ,
```
drug
```
  filters
- Output: guidelines with gene-drug pairs, dosing recommendations
```
fda_pharmacogenomic_biomarkers
```
-- FDA-approved PGx biomarkers
- Input: optional
```
drug_name
```
  ,
```
biomarker
```
  ,
```
limit
```
  (default 10; use
```
limit=1000
```
  for comprehensive results)
- Output:
```
{count, shown, results}
```
  with biomarker, drug, therapeutic area

Workflow:

Search CPIC for guidelines involving this drug
Query FDA PGx biomarkers with the drug name
For each PGx finding, note: the gene, the actionable alleles, and the clinical recommendation
Classify as: required testing (boxed warning), recommended testing, or informational

Tip: Use

limit=1000

with

fda_pharmacogenomic_biomarkers

to avoid missing entries (default limit is only 10).

目标: 识别影响药物安全性的遗传因素。

工具:

```
CPIC_list_guidelines
```
——获取CPIC药物基因组学指南
- Input: 可选
```
gene
```
  、
```
drug
```
  筛选条件
- Output: 包含基因-药物对、给药建议的指南列表
```
fda_pharmacogenomic_biomarkers
```
——FDA批准的PGx生物标志物
- Input: 可选
```
drug_name
```
  、
```
biomarker
```
  、
```
limit
```
  （默认10；使用
```
limit=1000
```
  获取全面结果）
- Output:
```
{count, shown, results}
```
  ，包含生物标志物、药物、治疗领域信息

工作流程:

搜索CPIC中涉及该药物的指南
使用药物名称查询FDA PGx生物标志物
对每个PGx发现，记录：基因、可操作等位基因及临床建议
分类为：需强制检测（黑框警告）、推荐检测或信息性内容

提示: 使用

fda_pharmacogenomic_biomarkers

时设置

limit=1000

，避免遗漏条目（默认限制仅为10）。

Phase 4: Clinical Trials

Phase 4: 临床试验

Objective: Find ongoing or completed trials studying drug safety.

Tools:

search_clinical_trials

-- search ClinicalTrials.gov

Input:
```
query_term
```
(required), optional
```
condition
```
,
```
intervention
```
,
```
pageSize
```
Output:
```
{studies, nextPageToken, total_count}
```
or string if no results

Workflow:

Search for safety-focused trials: "[drug] safety" or "[drug] adverse events"
Search for Risk Evaluation and Mitigation Strategies (REMS) trials
Look for post-marketing requirement (PMR) studies
Note trial status (recruiting, completed, terminated) and primary endpoints

Query tip: Simple queries work best. Complex multi-word queries often return no results. Search "[drug name]" first, then filter by safety-related keywords in the results.

目标: 查找进行中或已完成的药物安全性研究试验。

工具:

```
search_clinical_trials
```
——搜索ClinicalTrials.gov
- Input:
```
query_term
```
  （必填），可选
```
condition
```
  、
```
intervention
```
  、
```
pageSize
```
- Output:
```
{studies, nextPageToken, total_count}
```
  ；若无结果则返回字符串

工作流程:

搜索聚焦安全性的试验："[药物] safety"或"[药物] adverse events"
搜索风险评估与缓解策略（REMS）试验
查找上市后要求（PMR）研究
记录试验状态（招募中、已完成、终止）及主要终点

查询提示: 简单查询效果最佳。复杂的多词查询通常无结果。先搜索"[药物名称]"，再在结果中筛选与安全相关的关键词。

Phase 5: Literature Evidence

Phase 5: 文献证据

Objective: Find published safety studies, case reports, and meta-analyses.

Tools:

```
PubMed_search_articles
```
-- search biomedical literature
- Input:
```
query
```
  (search term), optional
```
limit
```
- Output: list of articles (plain list of dicts, NOT
```
{articles: [...]}
```
  )

Workflow:

Search: "[drug] adverse events" or "[drug] safety"
Search: "[drug] [specific adverse event]" for signals found in Phase 2
Look for systematic reviews and meta-analyses
Prioritize: meta-analyses > RCTs > cohort studies > case reports

目标: 查找已发表的安全性研究、病例报告及荟萃分析。

工具:

```
PubMed_search_articles
```
——搜索生物医学文献
- Input:
```
query
```
  （搜索词），可选
```
limit
```
- Output: 文章列表（纯字典列表，而非
```
{articles: [...]}
```
  ）

工作流程:

搜索："[药物] adverse events"或"[药物] safety"
针对Phase 2中发现的信号，搜索："[药物] [特定不良事件]"
查找系统评价及荟萃分析
优先级：荟萃分析 > 随机对照试验（RCT） > 队列研究 > 病例报告

Phase 6: Integrated Safety Report

Phase 6: 整合安全报告

Synthesize all phases into a cohesive report:

Drug Overview -- identity, class, mechanism, approval date, indications
Labeled Safety Information -- boxed warnings, key contraindications, known adverse reactions
Post-Market Signals -- FAERS signals with disproportionality metrics, compared to label
- Distinguish: known and labeled vs known but under-labeled vs potential new signal
Pharmacogenomic Considerations -- PGx biomarkers, testing recommendations
Clinical Trial Safety Data -- ongoing monitoring studies, REMS programs
Literature Summary -- key publications supporting or refining safety profile
Integrated Assessment -- overall risk characterization, populations at elevated risk, data gaps

Evidence grading:

T1: FDA label / regulatory action (boxed warning, REMS)
T2: Strong FAERS signal (PRR >= 5, multiple data sources agree)
T3: Moderate signal or single-source evidence
T4: Literature mention or computational prediction only

将所有阶段的内容整合为连贯的报告:

药物概述——识别信息、类别、作用机制、批准日期、适应症
标签载明的安全信息——黑框警告、关键禁忌症、已知不良反应
上市后信号——带不成比例性指标的FAERS信号，与标签内容对比
- 区分：已知且已载明 vs 已知但未充分载明 vs 潜在新信号
药物基因组学考量——PGx生物标志物、检测建议
临床试验安全数据——进行中的监测研究、REMS项目
文献摘要——支持或完善安全性特征的关键出版物
整合评估——整体风险特征、高风险人群、数据缺口

证据分级:

T1: FDA标签/监管行动（黑框警告、REMS）
T2: 强FAERS信号（PRR ≥ 5，多数据源一致）
T3: 中度信号或单一来源证据
T4: 仅文献提及或计算预测

Common Analysis Patterns

常见分析模式

Pattern	Description	Key Phases
Full Safety Review	Comprehensive regulatory-style review	All (0-6)
Label vs Real-World	Compare FDA label to FAERS signals	0, 1, 2, 6
PGx Safety Assessment	Focus on pharmacogenomic risk factors	0, 1, 3, 5
Signal Investigation	Deep-dive into a specific adverse event	0, 1, 2, 5, 6
Drug Comparison	Head-to-head safety comparison of two drugs	Run phases 0-2 for each, compare in Phase 6

模式	描述	核心阶段
全面安全审查	符合监管要求的综合性审查	全部（0-6）
标签 vs 真实世界	比较FDA标签与FAERS信号	0, 1, 2, 6
PGx安全性评估	聚焦药物基因组学风险因素	0, 1, 3, 5
信号调查	深入研究特定不良事件	0, 1, 2, 5, 6
药物对比	两种药物的安全性头对头比较	分别执行0-2阶段，在Phase 6中对比

Edge Cases & Fallbacks

边缘情况与 fallback 方案

New drug with little FAERS data: Rely on FDA label, clinical trials, and mechanism-based prediction
OTC drugs: May have limited FAERS data; DailyMed still has OTC labels
Combination products: Search FAERS for each active ingredient separately, then the combination
Brand vs generic discrepancies: FAERS reports may use either; search both forms
No CPIC guideline: Normal for most drugs; only ~30 gene-drug pairs have CPIC guidelines

FAERS数据较少的新药: 依赖FDA标签、临床试验及基于作用机制的预测
非处方（OTC）药物: FAERS数据可能有限；DailyMed仍提供OTC标签
复方产品: 分别搜索FAERS中每个活性成分的信息，再搜索复方产品
商品名与通用名差异: FAERS报告可能使用任意一种；两种形式都搜索
无CPIC指南: 多数药物均为正常情况；仅约30组基因-药物对有CPIC指南

Limitations

局限性

FAERS reporting bias: Spontaneous reports are voluntary; under-reporting is the norm
No denominator in FAERS: Cannot calculate incidence rates, only disproportionality
Label lag: FDA labels may not reflect the latest evidence; always supplement with FAERS and literature
PGx coverage: CPIC and FDA PGx biomarkers cover a fraction of all drugs
ClinicalTrials.gov completeness: Not all trials report results; some safety data is only in publications

FAERS报告偏倚: 自发报告为自愿提交；报告不足是常态
FAERS无分母数据: 无法计算发生率，仅能分析不成比例性
标签滞后: FDA标签可能未反映最新证据；需始终结合FAERS及文献补充
PGx覆盖范围有限: CPIC及FDA PGx生物标志物仅覆盖部分药物
ClinicalTrials.gov数据不完整: 并非所有试验都报告结果；部分安全数据仅存在于出版物中