tooluniverse-immunotherapy-response-prediction

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Immunotherapy Response Prediction

免疫治疗反应预测

Predict patient response to immune checkpoint inhibitors (ICIs) using multi-biomarker integration. Transforms a patient tumor profile (cancer type + mutations + biomarkers) into a quantitative ICI Response Score with drug-specific recommendations, resistance risk assessment, and monitoring plan.

KEY PRINCIPLES:

Report-first approach - Create report file FIRST, then populate progressively
Evidence-graded - Every finding has an evidence tier (T1-T4)
Quantitative output - ICI Response Score (0-100) with transparent component breakdown
Cancer-specific - All thresholds and predictions are cancer-type adjusted
Multi-biomarker - Integrate TMB + MSI + PD-L1 + neoantigen + mutations
Resistance-aware - Always check for known resistance mutations (STK11, PTEN, JAK1/2, B2M)
Drug-specific - Recommend specific ICI agents with evidence
Source-referenced - Every statement cites the tool/database source
English-first queries - Always use English terms in tool calls

通过多生物标志物整合预测患者对免疫检查点抑制剂（ICIs）的反应。将患者肿瘤特征（癌症类型+突变+生物标志物）转化为量化的ICI反应评分，同时提供药物特异性推荐、耐药风险评估及监测方案。

核心原则:

报告优先原则 - 先创建报告文件，再逐步填充内容
循证分级 - 所有结论均带有证据层级（T1-T4）
量化输出 - 提供ICI反应评分（0-100）及透明的成分拆解
癌症特异性 - 所有阈值与预测均针对癌症类型调整
多生物标志物 - 整合TMB + MSI + PD-L1 + 新抗原 + 突变数据
耐药性感知 - 始终检查已知耐药突变（STK11、PTEN、JAK1/2、B2M）
药物特异性 - 结合证据推荐特定ICI药物
来源溯源 - 所有陈述均标注工具/数据库来源
英文优先查询 - 工具调用中始终使用英文术语

When to Use

适用场景

Apply when user asks:

"Will this patient respond to immunotherapy?"
"Should I give pembrolizumab to this melanoma patient?"
"Patient has NSCLC with TMB 25, PD-L1 80% - predict ICI response"
"MSI-high colorectal cancer - which checkpoint inhibitor?"
"Patient has BRAF V600E melanoma, TMB 15 - immunotherapy or targeted?"
"Low TMB NSCLC with STK11 mutation - should I try immunotherapy?"
"Compare pembrolizumab vs nivolumab for this patient profile"
"What biomarkers predict checkpoint inhibitor response?"

当用户询问以下问题时适用：

"该患者对免疫治疗是否有反应？"
"我是否应为这名黑色素瘤患者使用pembrolizumab？"
"患者为NSCLC，TMB 25，PD-L1 80% - 预测ICI反应"
"MSI-H结直肠癌应选用哪种检查点抑制剂？"
"患者为BRAF V600E突变黑色素瘤，TMB 15 - 免疫治疗还是靶向治疗？"
"低TMB NSCLC伴STK11突变 - 能否尝试免疫治疗？"
"比较pembrolizumab与nivolumab对该患者的适用性"
"哪些生物标志物可预测检查点抑制剂反应？"

Input Parsing

输入解析

Required: Cancer type + at least one of: mutation list OR TMB value Optional: PD-L1 expression, MSI status, immune infiltration data, HLA type, prior treatments, intended ICI

必填项: 癌症类型 + 以下至少一项：突变列表或 TMB数值 可选项: PD-L1表达水平、MSI状态、免疫浸润数据、HLA分型、既往治疗史、拟用ICI药物

Accepted Input Formats

支持的输入格式

Format	Example	How to Parse
Cancer + mutations	"Melanoma, BRAF V600E, TP53 R273H"	cancer=melanoma, mutations=[BRAF V600E, TP53 R273H]
Cancer + TMB	"NSCLC, TMB 25 mut/Mb"	cancer=NSCLC, tmb=25
Cancer + full profile	"Melanoma, BRAF V600E, TMB 15, PD-L1 50%, MSS"	cancer=melanoma, mutations=[BRAF V600E], tmb=15, pdl1=50, msi=MSS
Cancer + MSI status	"Colorectal cancer, MSI-high"	cancer=CRC, msi=MSI-H
Resistance query	"NSCLC, TMB 2, STK11 loss, PD-L1 <1%"	cancer=NSCLC, tmb=2, mutations=[STK11 loss], pdl1=0
ICI selection	"Which ICI for NSCLC PD-L1 90%?"	cancer=NSCLC, pdl1=90, query_type=drug_selection

格式	示例	解析方式
癌症+突变	"Melanoma, BRAF V600E, TP53 R273H"	cancer=melanoma, mutations=[BRAF V600E, TP53 R273H]
癌症+TMB	"NSCLC, TMB 25 mut/Mb"	cancer=NSCLC, tmb=25
癌症+完整特征	"Melanoma, BRAF V600E, TMB 15, PD-L1 50%, MSS"	cancer=melanoma, mutations=[BRAF V600E], tmb=15, pdl1=50, msi=MSS
癌症+MSI状态	"Colorectal cancer, MSI-high"	cancer=CRC, msi=MSI-H
耐药性查询	"NSCLC, TMB 2, STK11 loss, PD-L1 <1%"	cancer=NSCLC, tmb=2, mutations=[STK11 loss], pdl1=0
ICI选择	"Which ICI for NSCLC PD-L1 90%?"	cancer=NSCLC, pdl1=90, query_type=drug_selection

Cancer Type Normalization

癌症类型标准化

Common aliases to resolve:

NSCLC -> non-small cell lung carcinoma
SCLC -> small cell lung carcinoma
CRC -> colorectal cancer
RCC -> renal cell carcinoma
HNSCC -> head and neck squamous cell carcinoma
UC / bladder -> urothelial carcinoma
HCC -> hepatocellular carcinoma
TNBC -> triple-negative breast cancer
GEJ -> gastroesophageal junction cancer

需解析的常见别名：

NSCLC -> 非小细胞肺癌
SCLC -> 小细胞肺癌
CRC -> 结直肠癌
RCC -> 肾细胞癌
HNSCC -> 头颈部鳞状细胞癌
UC / bladder -> 尿路上皮癌
HCC -> 肝细胞癌
TNBC -> 三阴性乳腺癌
GEJ -> 胃食管交界处癌

Gene Symbol Normalization

基因符号标准化

PD-L1 -> CD274
PD-1 -> PDCD1
CTLA-4 -> CTLA4
HER2 -> ERBB2
MSH2/MLH1/MSH6/PMS2 -> MMR genes

PD-L1 -> CD274
PD-1 -> PDCD1
CTLA-4 -> CTLA4
HER2 -> ERBB2
MSH2/MLH1/MSH6/PMS2 -> MMR基因

Phase 0: Tool Parameter Reference (CRITICAL)

阶段0：工具参数参考（至关重要）

BEFORE calling ANY tool, verify parameters using this reference table.

调用任何工具前，请使用此参考表验证参数。

Verified Tool Parameters

已验证的工具参数

Tool	Parameters	Notes
`OpenTargets_get_disease_id_description_by_name`	`diseaseName`	Returns `{data: {search: {hits: [{id, name, description}]}}}`
`OpenTargets_get_drug_id_description_by_name`	`drugName`	Returns `{data: {search: {hits: [{id, name, description}]}}}`
`OpenTargets_get_associated_drugs_by_disease_efoId`	`efoId` , `size`	Returns `{data: {disease: {knownDrugs: {count, rows}}}}`
`OpenTargets_get_drug_mechanisms_of_action_by_chemblId`	`chemblId`	Returns `{data: {drug: {mechanismsOfAction: {rows}}}}`
`OpenTargets_get_approved_indications_by_drug_chemblId`	`chemblId`	Approved indications list
`OpenTargets_get_drug_description_by_chemblId`	`chemblId`	Drug description text
`OpenTargets_get_associated_targets_by_drug_chemblId`	`chemblId`	Drug targets
`MyGene_query_genes`	`query` (NOT `q` )	Returns `{hits: [{_id, symbol, name, ensembl: {gene}}]}`
`ensembl_lookup_gene`	`gene_id` , `species='homo_sapiens'`	REQUIRES species. Returns `{data: {id, display_name}}`
`EnsemblVEP_annotate_rsid`	`variant_id` (NOT `rsid` )	VEP annotation with SIFT/PolyPhen
`civic_search_evidence_items`	`therapy_name` , `disease_name`	Returns `{data: {evidenceItems: {nodes}}}` - may not filter accurately
`civic_search_variants`	`name` , `gene_name`	Returns `{data: {variants: {nodes}}}` - returns many unrelated variants
`civic_get_variants_by_gene`	`gene_id` (CIViC numeric ID)	Requires CIViC gene ID, NOT Entrez
`civic_search_assertions`	`therapy_name` , `disease_name`	Returns `{data: {assertions: {nodes}}}`
`civic_search_therapies`	`name`	Search therapies by name
`cBioPortal_get_mutations`	`study_id` , `gene_list` (string)	`gene_list` is a STRING not array
`cBioPortal_get_cancer_studies`	(no params needed)	May fail with keyword param
`drugbank_get_drug_basic_info_by_drug_name_or_id`	`query` , `case_sensitive` , `exact_match` , `limit`	ALL 4 REQUIRED
`drugbank_get_targets_by_drug_name_or_drugbank_id`	`query` , `case_sensitive` , `exact_match` , `limit`	ALL 4 REQUIRED
`drugbank_get_pharmacology_by_drug_name_or_drugbank_id`	`query` , `case_sensitive` , `exact_match` , `limit`	ALL 4 REQUIRED
`drugbank_get_indications_by_drug_name_or_drugbank_id`	`query` , `case_sensitive` , `exact_match` , `limit`	ALL 4 REQUIRED
`FDA_get_indications_by_drug_name`	`drug_name` , `limit`	Returns `{meta, results}`
`FDA_get_clinical_studies_info_by_drug_name`	`drug_name` , `limit`	Returns `{meta, results}`
`FDA_get_adverse_reactions_by_drug_name`	`drug_name` , `limit`	Returns `{meta, results}`
`FDA_get_mechanism_of_action_by_drug_name`	`drug_name` , `limit`	Returns `{meta, results}`
`FDA_get_boxed_warning_info_by_drug_name`	`drug_name` , `limit`	May return NOT_FOUND
`FDA_get_warnings_by_drug_name`	`drug_name` , `limit`	Returns `{meta, results}`
`fda_pharmacogenomic_biomarkers`	`drug_name` , `biomarker` , `limit`	Returns `{count, shown, results: [{Drug, Biomarker, TherapeuticArea, LabelingSection}]}`
`clinical_trials_search`	`action='search_studies'` , `condition` , `intervention` , `limit`	Returns `{total_count, studies}`
`clinical_trials_get_details`	`action='get_study_details'` , `nct_id`	Full study object
`search_clinical_trials`	`query_term` (REQUIRED), `condition` , `intervention` , `pageSize`	Returns `{studies, total_count}`
`PubMed_search_articles`	`query` , `max_results`	Returns plain list of dicts
`UniProt_get_function_by_accession`	`accession`	Returns list of strings
`UniProt_get_disease_variants_by_accession`	`accession`	Disease-associated variants
`HPA_get_rna_expression_by_source`	`gene_name` , `source_type` , `source_name`	ALL 3 REQUIRED
`HPA_get_cancer_prognostics_by_gene`	`gene_name`	Cancer prognostic data
`iedb_search_epitopes`	`organism_name` , `source_antigen_name`	Returns `{status, data, count}`
`iedb_search_mhc`	various	MHC binding data
`enrichr_gene_enrichment_analysis`	`gene_list` (array), `libs` (array, REQUIRED)	Key libs: `KEGG_2021_Human` , `Reactome_2022`
`PharmGKB_get_clinical_annotations`	`query`	Clinical annotations
`gnomad_get_gene_constraints`	`gene_symbol`	Gene constraint metrics

工具	参数	说明
`OpenTargets_get_disease_id_description_by_name`	`diseaseName`	返回 `{data: {search: {hits: [{id, name, description}]}}}`
`OpenTargets_get_drug_id_description_by_name`	`drugName`	返回 `{data: {search: {hits: [{id, name, description}]}}}`
`OpenTargets_get_associated_drugs_by_disease_efoId`	`efoId` , `size`	返回 `{data: {disease: {knownDrugs: {count, rows}}}}`
`OpenTargets_get_drug_mechanisms_of_action_by_chemblId`	`chemblId`	返回 `{data: {drug: {mechanismsOfAction: {rows}}}}`
`OpenTargets_get_approved_indications_by_drug_chemblId`	`chemblId`	获批适应症列表
`OpenTargets_get_drug_description_by_chemblId`	`chemblId`	药物描述文本
`OpenTargets_get_associated_targets_by_drug_chemblId`	`chemblId`	药物靶点
`MyGene_query_genes`	`query` （非 `q` ）	返回 `{hits: [{_id, symbol, name, ensembl: {gene}}]}`
`ensembl_lookup_gene`	`gene_id` , `species='homo_sapiens'`	必须指定物种。返回 `{data: {id, display_name}}`
`EnsemblVEP_annotate_rsid`	`variant_id` （非 `rsid` ）	包含SIFT/PolyPhen的VEP注释
`civic_search_evidence_items`	`therapy_name` , `disease_name`	返回 `{data: {evidenceItems: {nodes}}}` - 过滤可能不准确
`civic_search_variants`	`name` , `gene_name`	返回 `{data: {variants: {nodes}}}` - 可能返回大量不相关变异
`civic_get_variants_by_gene`	`gene_id` （CIViC数值ID）	需要CIViC基因ID，而非Entrez ID
`civic_search_assertions`	`therapy_name` , `disease_name`	返回 `{data: {assertions: {nodes}}}`
`civic_search_therapies`	`name`	按名称搜索疗法
`cBioPortal_get_mutations`	`study_id` , `gene_list` （字符串）	`gene_list` 为字符串而非数组
`cBioPortal_get_cancer_studies`	（无参数）	若使用关键字参数可能失败
`drugbank_get_drug_basic_info_by_drug_name_or_id`	`query` , `case_sensitive` , `exact_match` , `limit`	四个参数均为必填
`drugbank_get_targets_by_drug_name_or_drugbank_id`	`query` , `case_sensitive` , `exact_match` , `limit`	四个参数均为必填
`drugbank_get_pharmacology_by_drug_name_or_drugbank_id`	`query` , `case_sensitive` , `exact_match` , `limit`	四个参数均为必填
`drugbank_get_indications_by_drug_name_or_drugbank_id`	`query` , `case_sensitive` , `exact_match` , `limit`	四个参数均为必填
`FDA_get_indications_by_drug_name`	`drug_name` , `limit`	返回 `{meta, results}`
`FDA_get_clinical_studies_info_by_drug_name`	`drug_name` , `limit`	返回 `{meta, results}`
`FDA_get_adverse_reactions_by_drug_name`	`drug_name` , `limit`	返回 `{meta, results}`
`FDA_get_mechanism_of_action_by_drug_name`	`drug_name` , `limit`	返回 `{meta, results}`
`FDA_get_boxed_warning_info_by_drug_name`	`drug_name` , `limit`	可能返回NOT_FOUND
`FDA_get_warnings_by_drug_name`	`drug_name` , `limit`	返回 `{meta, results}`
`fda_pharmacogenomic_biomarkers`	`drug_name` , `biomarker` , `limit`	返回 `{count, shown, results: [{Drug, Biomarker, TherapeuticArea, LabelingSection}]}`
`clinical_trials_search`	`action='search_studies'` , `condition` , `intervention` , `limit`	返回 `{total_count, studies}`
`clinical_trials_get_details`	`action='get_study_details'` , `nct_id`	完整研究对象
`search_clinical_trials`	`query_term` （必填）, `condition` , `intervention` , `pageSize`	返回 `{studies, total_count}`
`PubMed_search_articles`	`query` , `max_results`	返回普通字典列表
`UniProt_get_function_by_accession`	`accession`	返回字符串列表
`UniProt_get_disease_variants_by_accession`	`accession`	疾病相关变异
`HPA_get_rna_expression_by_source`	`gene_name` , `source_type` , `source_name`	三个参数均为必填
`HPA_get_cancer_prognostics_by_gene`	`gene_name`	癌症预后数据
`iedb_search_epitopes`	`organism_name` , `source_antigen_name`	返回 `{status, data, count}`
`iedb_search_mhc`	多参数	MHC结合数据
`enrichr_gene_enrichment_analysis`	`gene_list` （数组）, `libs` （数组，必填）	核心数据库： `KEGG_2021_Human` , `Reactome_2022`
`PharmGKB_get_clinical_annotations`	`query`	临床注释
`gnomad_get_gene_constraints`	`gene_symbol`	基因约束指标

Workflow Overview

工作流概览

Input: Cancer type + Mutations/TMB + Optional biomarkers (PD-L1, MSI, etc.)

Phase 1: Input Standardization & Cancer Context
  - Resolve cancer type to EFO ID
  - Parse mutation list
  - Resolve genes to Ensembl/Entrez IDs
  - Get cancer-specific ICI baseline

Phase 2: TMB Analysis
  - TMB classification (low/intermediate/high)
  - Cancer-specific TMB thresholds
  - FDA TMB-H biomarker status

Phase 3: Neoantigen Analysis
  - Estimate neoantigen burden from mutations
  - Mutation type classification (missense/frameshift/nonsense)
  - Neoantigen quality indicators

Phase 4: MSI/MMR Status Assessment
  - MSI status integration
  - MMR gene mutation check
  - FDA MSI-H approval status

Phase 5: PD-L1 Expression Analysis
  - PD-L1 level classification
  - Cancer-specific PD-L1 thresholds
  - FDA-approved PD-L1 cutoffs

Phase 6: Immune Microenvironment Profiling
  - Immune checkpoint gene expression
  - Tumor immune classification (hot/cold)
  - Immune escape signatures

Phase 7: Mutation-Based Predictors
  - Driver mutation analysis
  - Resistance mutations (STK11, PTEN, JAK1/2, B2M)
  - Sensitivity mutations (POLE)
  - DNA damage repair pathway

Phase 8: Clinical Evidence & ICI Options
  - FDA-approved ICIs for this cancer
  - Clinical trial response rates
  - Drug mechanism comparison
  - Combination therapy evidence

Phase 9: Resistance Risk Assessment
  - Known resistance factors
  - Tumor immune evasion mechanisms
  - Prior treatment context

Phase 10: Multi-Biomarker Score Integration
  - Calculate ICI Response Score (0-100)
  - Component breakdown
  - Confidence level

Phase 11: Clinical Recommendations
  - ICI drug recommendation
  - Monitoring plan
  - Alternative strategies

输入：癌症类型 + 突变/TMB + 可选生物标志物（PD-L1、MSI等）

阶段1：输入标准化与癌症背景
  - 解析癌症类型至EFO ID
  - 解析突变列表
  - 解析基因至Ensembl/Entrez ID
  - 获取癌症特异性ICI基线数据

阶段2：TMB分析
  - TMB分类（低/中/高）
  - 癌症特异性TMB阈值
  - FDA TMB-H生物标志物状态

阶段3：新抗原分析
  - 通过突变估算新抗原负荷
  - 突变类型分类（错义/移码/无义）
  - 新抗原质量指标

阶段4：MSI/MMR状态评估
  - 整合MSI状态
  - MMR基因突变检查
  - FDA MSI-H获批状态

阶段5：PD-L1表达分析
  - PD-L1水平分类
  - 癌症特异性PD-L1阈值
  - FDA获批PD-L1 cutoff值

阶段6：免疫微环境分析
  - 免疫检查点基因表达
  - 肿瘤免疫分类（热/冷）
  - 免疫逃逸特征

阶段7：基于突变的预测因子
  - 驱动突变分析
  - 耐药突变（STK11、PTEN、JAK1/2、B2M）
  - 敏感突变（POLE）
  - DNA损伤修复通路

阶段8：临床证据与ICI选项
  - 针对该癌症的FDA获批ICI药物
  - 临床试验反应率
  - 药物机制比较
  - 联合疗法证据

阶段9：耐药风险评估
  - 已知耐药因素
  - 肿瘤免疫逃逸机制
  - 既往治疗背景

阶段10：多生物标志物评分整合
  - 计算ICI反应评分（0-100）
  - 成分拆解
  - 置信水平

阶段11：临床推荐
  - ICI药物推荐
  - 监测方案
  - 替代策略

Phase 1: Input Standardization & Cancer Context

阶段1：输入标准化与癌症背景

Step 1.1: Resolve Cancer Type

步骤1.1：解析癌症类型

python

undefined

python

undefined

Get cancer EFO ID

获取癌症EFO ID

result = tu.tools.OpenTargets_get_disease_id_description_by_name(diseaseName='melanoma')

-> {data: {search: {hits: [{id: 'EFO_0000756', name: 'melanoma', description: '...'}]}}}


**Cancer-specific ICI context** (hardcoded knowledge base):

| Cancer Type | EFO ID | Baseline ICI ORR | Key Biomarkers | FDA-Approved ICIs |
|-------------|--------|-------------------|----------------|-------------------|
| Melanoma | EFO_0000756 | 30-45% | TMB, PD-L1 | pembro, nivo, ipi, nivo+ipi, nivo+rela |
| NSCLC | EFO_0003060 | 15-50% (PD-L1 dependent) | PD-L1, TMB, STK11 | pembro, nivo, atezo, durva, cemiplimab |
| Bladder/UC | EFO_0000292 | 15-25% | PD-L1, TMB | pembro, nivo, atezo, avelumab, durva |
| RCC | EFO_0000681 | 25-40% | PD-L1 | nivo, pembro, nivo+ipi, nivo+cabo, pembro+axitinib |
| HNSCC | EFO_0000181 | 15-20% | PD-L1 CPS | pembro, nivo |
| MSI-H (any) | N/A | 30-50% | MSI, dMMR | pembro (tissue-agnostic) |
| TMB-H (any) | N/A | 20-30% | TMB >=10 | pembro (tissue-agnostic) |
| CRC (MSI-H) | EFO_0000365 | 30-50% | MSI, dMMR | pembro, nivo, nivo+ipi |
| CRC (MSS) | EFO_0000365 | <5% | Generally poor | Generally not recommended |
| HCC | EFO_0000182 | 15-20% | PD-L1 | atezo+bev, durva+treme, nivo+ipi |
| TNBC | EFO_0005537 | 10-20% | PD-L1 CPS | pembro+chemo |
| Gastric/GEJ | EFO_0000178 | 10-20% | PD-L1 CPS, MSI | pembro, nivo |


**癌症特异性ICI背景**（硬编码知识库）：

| 癌症类型 | EFO ID | 基线ICI客观缓解率（ORR） | 关键生物标志物 | FDA获批ICI药物 |
|-------------|--------|-------------------|----------------|-------------------|
| 黑色素瘤 | EFO_0000756 | 30-45% | TMB、PD-L1 | pembro、nivo、ipi、nivo+ipi、nivo+rela |
| NSCLC | EFO_0003060 | 15-50%（依赖PD-L1） | PD-L1、TMB、STK11 | pembro、nivo、atezo、durva、cemiplimab |
| 膀胱癌/UC | EFO_0000292 | 15-25% | PD-L1、TMB | pembro、nivo、atezo、avelumab、durva |
| RCC | EFO_0000681 | 25-40% | PD-L1 | nivo、pembro、nivo+ipi、nivo+cabo、pembro+axitinib |
| HNSCC | EFO_0000181 | 15-20% | PD-L1 CPS | pembro、nivo |
| MSI-H（任意癌症） | N/A | 30-50% | MSI、dMMR | pembro（泛癌种获批） |
| TMB-H（任意癌症） | N/A | 20-30% | TMB >=10 | pembro（泛癌种获批） |
| CRC（MSI-H） | EFO_0000365 | 30-50% | MSI、dMMR | pembro、nivo、nivo+ipi |
| CRC（MSS） | EFO_0000365 | <5% | 整体反应差 | 通常不推荐 |
| HCC | EFO_0000182 | 15-20% | PD-L1 | atezo+bev、durva+treme、nivo+ipi |
| TNBC | EFO_0005537 | 10-20% | PD-L1 CPS | pembro+化疗 |
| 胃癌/GEJ | EFO_0000178 | 10-20% | PD-L1 CPS、MSI | pembro、nivo |

Step 1.2: Parse Mutations

步骤1.2：解析突变

Parse each mutation into structured format:

"BRAF V600E" -> {gene: "BRAF", variant: "V600E", type: "missense"}
"TP53 R273H" -> {gene: "TP53", variant: "R273H", type: "missense"}
"STK11 loss" -> {gene: "STK11", variant: "loss of function", type: "loss"}

将每个突变解析为结构化格式：

"BRAF V600E" -> {gene: "BRAF", variant: "V600E", type: "missense"}
"TP53 R273H" -> {gene: "TP53", variant: "R273H", type: "missense"}
"STK11 loss" -> {gene: "STK11", variant: "loss of function", type: "loss"}

Step 1.3: Resolve Gene IDs

步骤1.3：解析基因ID

python

undefined

python

undefined

For each gene in mutation list

针对突变列表中的每个基因

result = tu.tools.MyGene_query_genes(query='BRAF')

-> hits[0]: {_id: '673', symbol: 'BRAF', ensembl: {gene: 'ENSG00000157764'}}

---

---

Phase 2: TMB Analysis

阶段2：TMB分析

Step 2.1: TMB Classification

步骤2.1：TMB分类

If TMB value provided directly, classify:

TMB Range	Classification	ICI Score Component
>= 20 mut/Mb	TMB-High	30 points
10-19.9 mut/Mb	TMB-Intermediate	20 points
5-9.9 mut/Mb	TMB-Low	10 points
< 5 mut/Mb	TMB-Very-Low	5 points

If only mutations provided, estimate TMB:

Count total mutations provided
Note: User-provided lists are typically key mutations, not full exome
Flag as "estimated from provided mutations - clinical TMB testing recommended"

若直接提供TMB数值，分类如下：

TMB范围	分类	ICI评分成分
>=20 mut/Mb	TMB-High	30分
10-19.9 mut/Mb	TMB-Intermediate	20分
5-9.9 mut/Mb	TMB-Low	10分
<5 mut/Mb	TMB-Very-Low	5分

若仅提供突变列表，估算TMB：

统计提供的突变总数
注意：用户提供的列表通常为关键突变，而非全外显子组
标记为“基于提供的突变估算 - 建议进行临床TMB检测”

Step 2.2: TMB FDA Context

步骤2.2：TMB FDA背景

python

undefined

python

undefined

Check FDA TMB-H biomarker approval

检查FDA TMB-H生物标志物获批情况

result = tu.tools.fda_pharmacogenomic_biomarkers(drug_name='pembrolizumab', limit=100)

Look for "Tumor Mutational Burden" in Biomarker field

在Biomarker字段中查找"Tumor Mutational Burden"

-> Pembrolizumab approved for TMB-H (>=10 mut/Mb) tissue-agnostic

-> Pembrolizumab获批用于TMB-H（>=10 mut/Mb）泛癌种

undefined

undefined

Step 2.3: Cancer-Specific TMB Thresholds

步骤2.3：癌症特异性TMB阈值

Cancer Type	Typical TMB Range	High-TMB Threshold	Notes
Melanoma	5-50+	>20	High baseline TMB; UV-induced
NSCLC	2-30	>10	Smoking-related; FDA cutoff 10
Bladder	5-25	>10	Moderate baseline
CRC (MSI-H)	20-100+	>10	Very high in MSI-H
CRC (MSS)	2-10	>10	Generally low
RCC	1-8	>10	Low TMB but ICI-responsive
HNSCC	2-15	>10	Moderate

IMPORTANT: RCC responds to ICIs despite low TMB. TMB is less predictive in some cancers.

癌症类型	典型TMB范围	高TMB阈值	说明
黑色素瘤	5-50+	>20	基线TMB高；紫外线诱导
NSCLC	2-30	>10	吸烟相关；FDA cutoff值为10
膀胱癌	5-25	>10	基线中等
CRC（MSI-H）	20-100+	>10	MSI-H患者TMB极高
CRC（MSS）	2-10	>10	通常较低
RCC	1-8	>10	TMB低但对ICI有反应
HNSCC	2-15	>10	中等水平

重要提示：RCC尽管TMB低，但对ICI有反应。TMB在部分癌症中的预测性较差。

Phase 3: Neoantigen Analysis

阶段3：新抗原分析

Step 3.1: Neoantigen Burden Estimation

步骤3.1：新抗原负荷估算

From mutation list:

Missense mutations -> Each has ~20-50% chance of generating a neoantigen
Frameshift mutations -> High neoantigen-generating potential (novel peptides)
Nonsense mutations -> Moderate potential (truncated proteins)
Splice site mutations -> Moderate potential (aberrant peptides)

Estimate: neoantigen_count ~= missense_count * 0.3 + frameshift_count * 1.5

基于突变列表：

错义突变 -> 每个突变约有20-50%的概率产生新抗原
移码突变 -> 新抗原产生潜力高（新型肽段）
无义突变 -> 中等潜力（截短蛋白）
剪接位点突变 -> 中等潜力（异常肽段）

估算公式：新抗原数量 ≈ 错义突变数 * 0.3 + 移码突变数 * 1.5

Step 3.2: Neoantigen Quality Assessment

步骤3.2：新抗原质量评估

python

undefined

python

undefined

Check mutation impact using UniProt

使用UniProt检查突变影响

result = tu.tools.UniProt_get_function_by_accession(accession='P15056') # BRAF UniProt

result = tu.tools.UniProt_get_function_by_accession(accession='P15056') # BRAF的UniProt编号

Assess if mutation is in functional domain

评估突变是否位于功能结构域


**Quality indicators**:
- Mutations in protein kinase domains -> high immunogenicity potential
- Mutations in surface-exposed regions -> better MHC presentation
- POLE/POLD1 mutations -> ultra-high neoantigen load (ultramutated)


**质量指标**:
- 位于蛋白激酶结构域的突变 -> 免疫原性潜力高
- 位于表面暴露区域的突变 -> MHC呈递效果更好
- POLE/POLD1突变 -> 超高新抗原负荷（超突变）

Step 3.3: IEDB Epitope Data (if relevant)

步骤3.3：IEDB表位数据（如适用）

python

undefined

python

undefined

Check known epitopes for mutated proteins

检查突变蛋白的已知表位

result = tu.tools.iedb_search_epitopes(organism_name='homo sapiens', source_antigen_name='BRAF')

Returns known epitopes, MHC restrictions

返回已知表位、MHC限制性

undefined

undefined

Neoantigen Score Component

新抗原评分成分

Estimated Neoantigen Load	Classification	Score
>50 neoantigens	High	15 points
20-50 neoantigens	Moderate	10 points
<20 neoantigens	Low	5 points

估算新抗原负荷	分类	评分
>50个新抗原	高	15分
20-50个新抗原	中	10分
<20个新抗原	低	5分

Phase 4: MSI/MMR Status Assessment

阶段4：MSI/MMR状态评估

Step 4.1: MSI Status Integration

步骤4.1：整合MSI状态

If MSI status provided directly:

MSI Status	Classification	Score Component
MSI-H / dMMR	MSI-High	25 points
MSS / pMMR	Microsatellite Stable	5 points
Unknown	Not tested	10 points (neutral)

若直接提供MSI状态：

MSI状态	分类	评分成分
MSI-H / dMMR	MSI-High	25分
MSS / pMMR	微卫星稳定	5分
未知	未检测	10分（中性）

Step 4.2: MMR Gene Mutation Check

步骤4.2：MMR基因突变检查

Check if any provided mutations are in MMR genes:

MLH1 (ENSG00000076242) - mismatch repair
MSH2 (ENSG00000095002) - mismatch repair
MSH6 (ENSG00000116062) - mismatch repair
PMS2 (ENSG00000122512) - mismatch repair
EPCAM (ENSG00000119888) - can silence MSH2

If MMR gene mutations found but MSI status not provided -> flag as "possible MSI-H, recommend testing"

检查提供的突变是否位于MMR基因：

MLH1（ENSG00000076242） - 错配修复
MSH2（ENSG00000095002） - 错配修复
MSH6（ENSG00000116062） - 错配修复
PMS2（ENSG00000122512） - 错配修复
EPCAM（ENSG00000119888） - 可沉默MSH2

若发现MMR基因突变但未提供MSI状态 -> 标记为“可能为MSI-H，建议检测”

Step 4.3: FDA MSI-H Approvals

步骤4.3：FDA MSI-H获批情况

python

undefined

python

undefined

Check FDA approvals for MSI-H

检查MSI-H的FDA获批情况

result = tu.tools.fda_pharmacogenomic_biomarkers(biomarker='Microsatellite Instability', limit=100)

Pembrolizumab: tissue-agnostic for MSI-H/dMMR

Pembrolizumab：泛癌种获批用于MSI-H/dMMR

Nivolumab: CRC (MSI-H)

Nivolumab：获批用于MSI-H结直肠癌

Dostarlimab: dMMR solid tumors

Dostarlimab：获批用于dMMR实体瘤

---

---

Phase 5: PD-L1 Expression Analysis

阶段5：PD-L1表达分析

Step 5.1: PD-L1 Level Classification

步骤5.1：PD-L1水平分类

PD-L1 Level	Classification	Score Component
>= 50% (TPS)	PD-L1 High	20 points
1-49% (TPS)	PD-L1 Positive	12 points
< 1% (TPS)	PD-L1 Negative	5 points
Unknown	Not tested	10 points (neutral)

PD-L1水平	分类	评分成分
>=50%（TPS）	PD-L1 High	20分
1-49%（TPS）	PD-L1 Positive	12分
<1%（TPS）	PD-L1 Negative	5分
未知	未检测	10分（中性）

Step 5.2: Cancer-Specific PD-L1 Thresholds

步骤5.2：癌症特异性PD-L1阈值

Cancer	Scoring Method	Key Thresholds	ICI Monotherapy Recommended?
NSCLC	TPS	>=50%: first-line mono; >=1%: after chemo	Yes at >=50%, combo at >=1%
Melanoma	Not routinely required	N/A	Yes regardless of PD-L1
Bladder	CPS or IC	CPS>=10 preferred	Yes with PD-L1 positive
HNSCC	CPS	CPS>=1: pembro; CPS>=20: mono preferred	CPS>=20 for monotherapy
Gastric	CPS	CPS>=1	Pembro+chemo
TNBC	CPS	CPS>=10	Pembro+chemo

癌症	评分方法	关键阈值	是否推荐ICI单药治疗？
NSCLC	TPS	>=50%：一线单药；>=1%：化疗后	>=50%推荐单药，>=1%推荐联合
黑色素瘤	常规无需检测	N/A	无论PD-L1水平均推荐
膀胱癌	CPS或IC	优先CPS>=10	PD-L1阳性时推荐
HNSCC	CPS	CPS>=1：使用pembro；CPS>=20：优先单药	CPS>=20推荐单药
胃癌	CPS	CPS>=1	pembro+化疗
TNBC	CPS	CPS>=10	pembro+化疗

Step 5.3: PD-L1 Gene Expression (Baseline Reference)

步骤5.3：PD-L1基因表达（基线参考）

python

undefined

python

undefined

PD-L1 (CD274) expression patterns

PD-L1（CD274）表达模式

result = tu.tools.HPA_get_cancer_prognostics_by_gene(gene_name='CD274')

Cancer-type specific prognostic data

癌症特异性预后数据

---

---

Phase 6: Immune Microenvironment Profiling

阶段6：免疫微环境分析

Step 6.1: Key Immune Checkpoint Genes

步骤6.1：关键免疫检查点基因

Query expression data for immune microenvironment markers:

python

undefined

查询免疫微环境标记物的表达数据：

python

undefined

Key immune genes to check

需检查的关键免疫基因

immune_genes = ['CD274', 'PDCD1', 'CTLA4', 'LAG3', 'HAVCR2', 'TIGIT', 'CD8A', 'CD8B', 'GZMA', 'GZMB', 'PRF1', 'IFNG']

For each gene, get cancer-specific expression

针对每个基因，获取癌症特异性表达数据

for gene in immune_genes: result = tu.tools.HPA_get_cancer_prognostics_by_gene(gene_name=gene)

undefined

for gene in immune_genes: result = tu.tools.HPA_get_cancer_prognostics_by_gene(gene_name=gene)

undefined

Step 6.2: Tumor Immune Classification

步骤6.2：肿瘤免疫分类

Based on available data, classify:

Classification	Characteristics	ICI Likelihood
Hot (T cell inflamed)	High CD8+ T cells, IFN-g, PD-L1+	High response
Cold (immune desert)	Low immune infiltration	Low response
Immune excluded	Immune cells at margin, not infiltrating	Moderate response
Immune suppressed	High Tregs, MDSCs, immunosuppressive	Low-moderate

基于可用数据分类：

分类	特征	ICI反应可能性
热肿瘤（T细胞浸润）	高CD8+ T细胞、IFN-g、PD-L1+	高反应
冷肿瘤（免疫荒漠）	低免疫浸润	低反应
免疫排斥型	免疫细胞位于肿瘤边缘，未浸润	中等反应
免疫抑制型	高Tregs、MDSCs、免疫抑制因子	低-中等反应

Step 6.3: Immune Pathway Enrichment

步骤6.3：免疫通路富集分析

python

undefined

python

undefined

If mutation list includes immune-related genes, do pathway analysis

若突变列表包含免疫相关基因，进行通路分析

result = tu.tools.enrichr_gene_enrichment_analysis( gene_list=['CD274', 'PDCD1', 'CTLA4', 'IFNG', 'CD8A'], libs=['KEGG_2021_Human', 'Reactome_2022'] )

---

result = tu.tools.enrichr_gene_enrichment_analysis( gene_list=['CD274', 'PDCD1', 'CTLA4', 'IFNG', 'CD8A'], libs=['KEGG_2021_Human', 'Reactome_2022'] )

---

Phase 7: Mutation-Based Predictors

阶段7：基于突变的预测因子

Step 7.1: ICI-Resistance Mutations (CRITICAL)

步骤7.1：ICI耐药突变（至关重要）

Known resistance mutations - apply PENALTIES:

Gene	Mutation	Cancer Context	Mechanism	Penalty
STK11/LKB1	Loss/inactivation	NSCLC (esp. KRAS+)	Immune exclusion, cold TME	-10 points
PTEN	Loss/deletion	Multiple	Reduced T cell infiltration	-5 points
JAK1	Loss of function	Multiple	IFN-g signaling loss	-10 points
JAK2	Loss of function	Multiple	IFN-g signaling loss	-10 points
B2M	Loss/mutation	Multiple	MHC-I loss, immune escape	-15 points
KEAP1	Loss/mutation	NSCLC	Oxidative stress, cold TME	-5 points
MDM2	Amplification	Multiple	Hyperprogression risk	-5 points
MDM4	Amplification	Multiple	Hyperprogression risk	-5 points
EGFR	Activating mutation	NSCLC	Low TMB, cold TME	-5 points

已知耐药突变 - 应用扣分：

基因	突变	癌症背景	机制	扣分
STK11/LKB1	缺失/失活	NSCLC（尤其KRAS+）	免疫排斥、冷肿瘤微环境	-10分
PTEN	缺失/删除	多种癌症	T细胞浸润减少	-5分
JAK1	功能缺失	多种癌症	IFN-g信号通路丢失	-10分
JAK2	功能缺失	多种癌症	IFN-g信号通路丢失	-10分
B2M	缺失/突变	多种癌症	MHC-I丢失、免疫逃逸	-15分
KEAP1	缺失/突变	NSCLC	氧化应激、冷肿瘤微环境	-5分
MDM2	扩增	多种癌症	超进展风险	-5分
MDM4	扩增	多种癌症	超进展风险	-5分
EGFR	激活突变	NSCLC	低TMB、冷肿瘤微环境	-5分

Step 7.2: ICI-Sensitivity Mutations (BONUS)

步骤7.2：ICI敏感突变（加分）

Gene	Mutation	Cancer Context	Mechanism	Bonus
POLE	Exonuclease domain	Any	Ultramutation, high neoantigens	+10 points
POLD1	Proofreading domain	Any	Ultramutation	+5 points
BRCA1/2	Loss of function	Multiple	Genomic instability	+3 points
ARID1A	Loss of function	Multiple	Chromatin remodeling, TME	+3 points
PBRM1	Loss of function	RCC	ICI response in RCC	+5 points (RCC only)

基因	突变	癌症背景	机制	加分
POLE	核酸酶结构域突变	任意癌症	超突变、高新抗原负荷	+10分
POLD1	校对结构域突变	任意癌症	超突变	+5分
BRCA1/2	功能缺失	多种癌症	基因组不稳定性	+3分
ARID1A	功能缺失	多种癌症	染色质重塑、肿瘤微环境	+3分
PBRM1	功能缺失	RCC	RCC对ICI有反应	+5分（仅RCC）

Step 7.3: Driver Mutation Context

步骤7.3：驱动突变背景

python

undefined

python

undefined

For each mutation, check CIViC evidence for ICI context

针对每个突变，检查CIViC中与ICI相关的证据

Use OpenTargets for drug associations

使用OpenTargets获取药物关联

result = tu.tools.OpenTargets_get_associated_drugs_by_disease_efoId(efoId='EFO_0000756', size=50)

Filter for ICI drugs (pembro, nivo, ipi, atezo, durva, avelumab, cemiplimab)

筛选ICI药物（pembro、nivo、ipi、atezo、durva、avelumab、cemiplimab）

undefined

undefined

Step 7.4: DNA Damage Repair (DDR) Pathway

步骤7.4：DNA损伤修复（DDR）通路

Check if mutations are in DDR genes (associated with ICI response):

ATM, ATR, CHEK1, CHEK2 - DNA damage sensing
BRCA1, BRCA2, PALB2 - homologous recombination
RAD50, MRE11, NBN - double-strand break repair
POLE, POLD1 - polymerase proofreading

DDR mutations -> likely higher TMB -> better ICI response

检查突变是否位于与ICI反应相关的DDR基因：

ATM、ATR、CHEK1、CHEK2 - DNA损伤感知
BRCA1、BRCA2、PALB2 - 同源重组
RAD50、MRE11、NBN - 双链断裂修复
POLE、POLD1 - 聚合酶校对

DDR突变 -> 通常TMB更高 -> ICI反应更好

Phase 8: Clinical Evidence & ICI Options

阶段8：临床证据与ICI选项

Step 8.1: FDA-Approved ICIs

步骤8.1：FDA获批ICI药物

python

undefined

python

undefined

Get FDA indications for key ICIs

获取关键ICI药物的FDA适应症

ici_drugs = ['pembrolizumab', 'nivolumab', 'atezolizumab', 'durvalumab', 'ipilimumab', 'avelumab', 'cemiplimab']

for drug in ici_drugs: result = tu.tools.FDA_get_indications_by_drug_name(drug_name=drug, limit=3) # Extract cancer-specific indications

undefined

ici_drugs = ['pembrolizumab', 'nivolumab', 'atezolizumab', 'durvalumab', 'ipilimumab', 'avelumab', 'cemiplimab']

for drug in ici_drugs: result = tu.tools.FDA_get_indications_by_drug_name(drug_name=drug, limit=3) # 提取癌症特异性适应症

undefined

Step 8.2: ICI Drug Profiles

步骤8.2：ICI药物概况

Drug	Target	Type	Key Indications
Pembrolizumab (Keytruda)	PD-1	IgG4 mAb	Melanoma, NSCLC, HNSCC, Bladder, MSI-H, TMB-H, many others
Nivolumab (Opdivo)	PD-1	IgG4 mAb	Melanoma, NSCLC, RCC, CRC (MSI-H), HCC, HNSCC
Atezolizumab (Tecentriq)	PD-L1	IgG1 mAb	NSCLC, Bladder, HCC, Melanoma
Durvalumab (Imfinzi)	PD-L1	IgG1 mAb	NSCLC (Stage III), Bladder, HCC, BTC
Ipilimumab (Yervoy)	CTLA-4	IgG1 mAb	Melanoma, RCC (combo), CRC (MSI-H combo)
Avelumab (Bavencio)	PD-L1	IgG1 mAb	Merkel cell, Bladder (maintenance)
Cemiplimab (Libtayo)	PD-1	IgG4 mAb	CSCC, NSCLC, Basal cell
Dostarlimab (Jemperli)	PD-1	IgG4 mAb	dMMR endometrial, dMMR solid tumors
Tremelimumab (Imjudo)	CTLA-4	IgG2 mAb	HCC (combo with durva)

药物	靶点	类型	关键适应症
Pembrolizumab（Keytruda）	PD-1	IgG4单抗	黑色素瘤、NSCLC、HNSCC、膀胱癌、MSI-H、TMB-H等多种癌症
Nivolumab（Opdivo）	PD-1	IgG4单抗	黑色素瘤、NSCLC、RCC、MSI-H结直肠癌、HCC、HNSCC
Atezolizumab（Tecentriq）	PD-L1	IgG1单抗	NSCLC、膀胱癌、HCC、黑色素瘤
Durvalumab（Imfinzi）	PD-L1	IgG1单抗	NSCLC（III期）、膀胱癌、HCC、BTC
Ipilimumab（Yervoy）	CTLA-4	IgG1单抗	黑色素瘤、RCC（联合）、MSI-H结直肠癌（联合）
Avelumab（Bavencio）	PD-L1	IgG1单抗	默克尔细胞癌、膀胱癌（维持治疗）
Cemiplimab（Libtayo）	PD-1	IgG4单抗	CSCC、NSCLC、基底细胞癌
Dostarlimab（Jemperli）	PD-1	IgG4单抗	dMMR子宫内膜癌、dMMR实体瘤
Tremelimumab（Imjudo）	CTLA-4	IgG2单抗	HCC（与durva联合）

Step 8.3: Clinical Trial Evidence

步骤8.3：临床试验证据

python

undefined

python

undefined

Search for ICI trials in this cancer type

搜索该癌症类型的ICI临床试验

result = tu.tools.clinical_trials_search( action='search_studies', condition='melanoma', intervention='pembrolizumab', limit=10 )

Returns: {total_count, studies: [{nctId, title, status, conditions}]}

返回：{total_count, studies: [{nctId, title, status, conditions}]}

undefined

undefined

Step 8.4: Literature Evidence

步骤8.4：文献证据

python

undefined

python

undefined

Search PubMed for biomarker-specific ICI response data

搜索PubMed中生物标志物特异性ICI反应数据

result = tu.tools.PubMed_search_articles( query='pembrolizumab melanoma TMB response biomarker', max_results=10 )

Returns list of {pmid, title, ...}

返回{pmid, title, ...}列表

undefined

undefined

Step 8.5: OpenTargets Drug-Target Evidence

步骤8.5：OpenTargets药物-靶点证据

python

undefined

python

undefined

Get drug mechanism details

获取药物机制细节

result = tu.tools.OpenTargets_get_drug_mechanisms_of_action_by_chemblId(chemblId='CHEMBL3137343')

-> pembrolizumab: PD-1 inhibitor, targets PDCD1 (ENSG00000188389)

-> pembrolizumab：PD-1抑制剂，靶点为PDCD1（ENSG00000188389）

undefined

undefined

Key ICI ChEMBL IDs

关键ICI的ChEMBL ID

Drug	ChEMBL ID
Pembrolizumab	CHEMBL3137343
Nivolumab	CHEMBL2108738
Atezolizumab	CHEMBL3707227
Durvalumab	CHEMBL3301587
Ipilimumab	CHEMBL1789844
Avelumab	CHEMBL3833373
Cemiplimab	CHEMBL4297723

药物	ChEMBL ID
Pembrolizumab	CHEMBL3137343
Nivolumab	CHEMBL2108738
Atezolizumab	CHEMBL3707227
Durvalumab	CHEMBL3301587
Ipilimumab	CHEMBL1789844
Avelumab	CHEMBL3833373
Cemiplimab	CHEMBL4297723

Phase 9: Resistance Risk Assessment

阶段9：耐药风险评估

Step 9.1: Known Resistance Factors Check

步骤9.1：已知耐药因素检查

For each mutation in the patient profile, check against resistance database:

python

undefined

针对患者特征中的每个突变，检查耐药数据库：

python

undefined

Check for resistance evidence in CIViC

检查CIViC中与pembrolizumab相关的耐药证据

CIViC evidence types: PREDICTIVE, PROGNOSTIC, DIAGNOSTIC, PREDISPOSING, ONCOGENIC

CIViC证据类型：PREDICTIVE、PROGNOSTIC、DIAGNOSTIC、PREDISPOSING、ONCOGENIC

result = tu.tools.civic_search_evidence_items(therapy_name='pembrolizumab')

Filter for resistance-associated evidence

筛选与耐药相关的证据

undefined

undefined

Step 9.2: Pathway-Level Resistance

步骤9.2：通路水平耐药

Pathway	Resistance Mechanism	Genes
IFN-g signaling	Loss of IFN-g response	JAK1, JAK2, STAT1, IRF1
Antigen presentation	MHC-I downregulation	B2M, TAP1, TAP2, HLA-A/B/C
WNT/b-catenin	T cell exclusion	CTNNB1 activating mutations
MAPK pathway	Immune suppression	MEK, ERK hyperactivation
PI3K/AKT/mTOR	Immune suppression	PTEN loss, PIK3CA

通路	耐药机制	基因
IFN-g信号通路	IFN-g反应丢失	JAK1、JAK2、STAT1、IRF1
抗原呈递	MHC-I下调	B2M、TAP1、TAP2、HLA-A/B/C
WNT/β-catenin	T细胞排斥	CTNNB1激活突变
MAPK通路	免疫抑制	MEK、ERK过度激活
PI3K/AKT/mTOR	免疫抑制	PTEN缺失、PIK3CA突变

Step 9.3: Resistance Risk Score

步骤9.3：耐药风险评分

Summarize resistance risk as:

Low risk: No resistance mutations, favorable TME
Moderate risk: 1 resistance factor OR uncertain TME
High risk: Multiple resistance mutations OR known resistant phenotype

总结耐药风险：

低风险：无耐药突变，肿瘤微环境有利
中风险：1项耐药因素或肿瘤微环境不确定
高风险：多项耐药突变或已知耐药表型

Phase 10: Multi-Biomarker Score Integration

阶段10：多生物标志物评分整合

ICI Response Score Calculation (0-100)

ICI反应评分计算（0-100）

TOTAL SCORE = TMB_score + MSI_score + PDL1_score + Neoantigen_score + Mutation_bonus + Resistance_penalty

Where:
  TMB_score:        5-30 points (based on TMB classification)
  MSI_score:        5-25 points (based on MSI status)
  PDL1_score:       5-20 points (based on PD-L1 level)
  Neoantigen_score: 5-15 points (based on estimated neoantigens)
  Mutation_bonus:   0-10 points (POLE, PBRM1, etc.)
  Resistance_penalty: -20 to 0 points (STK11, PTEN, JAK1/2, B2M)

Minimum score: 0 (floor)
Maximum score: 100 (cap)

总评分 = TMB评分 + MSI评分 + PD-L1评分 + 新抗原评分 + 突变加分 + 耐药扣分

其中：
  TMB评分:        5-30分（基于TMB分类）
  MSI评分:        5-25分（基于MSI状态）
  PD-L1评分:       5-20分（基于PD-L1水平）
  新抗原评分: 5-15分（基于估算新抗原数量）
  突变加分:   0-10分（POLE、PBRM1等）
  耐药扣分: -20至0分（STK11、PTEN、JAK1/2、B2M）

最低分: 0（下限）
最高分: 100（上限）

Response Likelihood Tiers

反应可能性层级

Score Range	Tier	Expected ORR	Recommendation
70-100	HIGH	50-80%	Strong ICI candidate; monotherapy or combo
40-69	MODERATE	20-50%	Consider ICI; combo preferred; monitor closely
0-39	LOW	<20%	ICI alone unlikely effective; consider alternatives

评分范围	层级	预期ORR	推荐
70-100	高	50-80%	强烈推荐ICI；单药或联合治疗
40-69	中	20-50%	考虑ICI；优先联合治疗；密切监测
0-39	低	<20%	ICI单药可能无效；考虑替代方案

Confidence Level

置信水平

Data Completeness	Confidence
All biomarkers (TMB + MSI + PD-L1 + mutations)	HIGH
3 of 4 biomarkers	MODERATE-HIGH
2 of 4 biomarkers	MODERATE
1 biomarker only	LOW
Cancer type only	VERY LOW

数据完整性	置信度
所有生物标志物（TMB + MSI + PD-L1 + 突变）	高
4项中3项生物标志物	中-高
4项中2项生物标志物	中
仅1项生物标志物	低
仅癌症类型	极低

Phase 11: Clinical Recommendations

阶段11：临床推荐

Step 11.1: ICI Drug Selection Algorithm

步骤11.1：ICI药物选择算法

IF MSI-H:
  -> Pembrolizumab (tissue-agnostic FDA approval)
  -> Nivolumab (CRC-specific)
  -> Consider nivo+ipi combination

IF TMB-H (>=10) and not MSI-H:
  -> Pembrolizumab (tissue-agnostic for TMB-H)

IF Cancer = Melanoma:
  IF PD-L1 >= 1%: pembrolizumab or nivolumab monotherapy
  ELSE: nivolumab + ipilimumab combination
  IF BRAF V600E: consider targeted therapy first if rapid response needed

IF Cancer = NSCLC:
  IF PD-L1 >= 50% and no STK11/EGFR: pembrolizumab monotherapy
  IF PD-L1 1-49%: pembrolizumab + chemotherapy
  IF PD-L1 < 1%: ICI + chemotherapy combination
  IF STK11 loss: ICI less likely effective
  IF EGFR/ALK positive: targeted therapy preferred over ICI

IF Cancer = RCC:
  -> Nivolumab + ipilimumab (IMDC intermediate/poor risk)
  -> Pembrolizumab + axitinib (all risk)

IF Cancer = Bladder:
  -> Pembrolizumab or atezolizumab (2L)
  -> Avelumab maintenance post-platinum

若为MSI-H:
  -> Pembrolizumab（泛癌种FDA获批）
  -> Nivolumab（结直肠癌特异性）
  -> 考虑nivo+ipi联合治疗

若为TMB-H（>=10）且非MSI-H:
  -> Pembrolizumab（泛癌种TMB-H获批）

若癌症为黑色素瘤:
  若PD-L1 >=1%: pembrolizumab或nivolumab单药
  否则: nivolumab + ipilimumab联合治疗
  若为BRAF V600E突变: 若需快速缓解，考虑先使用靶向治疗

若癌症为NSCLC:
  若PD-L1 >=50%且无STK11/EGFR突变: pembrolizumab单药
  若PD-L1 1-49%: pembrolizumab + 化疗
  若PD-L1 <1%: ICI + 化疗联合
  若存在STK11缺失: ICI可能无效
  若EGFR/ALK阳性: 优先靶向治疗而非ICI

若癌症为RCC:
  -> Nivolumab + ipilimumab（IMDC中/低风险）
  -> Pembrolizumab + axitinib（所有风险分层）

若癌症为膀胱癌:
  -> Pembrolizumab或atezolizumab（二线）
  -> Avelumab铂类治疗后维持

Step 11.2: Monitoring Plan

步骤11.2：监测方案

During ICI treatment, monitor:

Tumor response (CT/MRI every 8-12 weeks)
Circulating tumor DNA (ctDNA) for early response
Immune-related adverse events (irAEs)
Thyroid function (TSH every 6 weeks)
Liver function (every 2-4 weeks initially)
Cortisol if symptoms

Early response biomarkers:

ctDNA decrease at 4-6 weeks
PET-CT metabolic response
Circulating immune cell phenotyping

ICI治疗期间，需监测:

肿瘤反应（每8-12周进行CT/MRI）
循环肿瘤DNA（ctDNA）以早期评估反应
免疫相关不良事件（irAEs）
甲状腺功能（每6周检测TSH）
肝功能（初始每2-4周检测）
若有症状检测皮质醇

早期反应生物标志物:

4-6周时ctDNA下降
PET-CT代谢反应
循环免疫细胞表型分析

Step 11.3: Alternative Strategies

步骤11.3：替代策略

If ICI response predicted to be LOW:

Targeted therapy (if actionable mutations: BRAF, EGFR, ALK, ROS1)
Chemotherapy (standard of care)
ICI + chemotherapy combination (may overcome low PD-L1)
ICI + anti-angiogenic (may convert cold to hot tumor)
ICI + CTLA-4 combo (nivolumab + ipilimumab)
Clinical trial enrollment (novel combinations)

若预测ICI反应为低:

靶向治疗（若存在可靶向突变：BRAF、EGFR、ALK、ROS1）
化疗（标准治疗）
ICI + 化疗联合（可能克服低PD-L1）
ICI + 抗血管生成药物（可能将冷肿瘤转化为热肿瘤）
ICI + CTLA-4联合（nivolumab + ipilimumab）
参与临床试验（新型联合疗法）

Output Report Format

输出报告格式

Save report as

immunotherapy_response_prediction_{cancer_type}.md

将报告保存为

immunotherapy_response_prediction_{cancer_type}.md

Report Structure

报告结构

markdown

undefined

markdown

undefined

Immunotherapy Response Prediction Report

免疫治疗反应预测报告

Executive Summary

执行摘要

[2-3 sentence summary: cancer type, ICI Response Score, recommendation]

[2-3句总结：癌症类型、ICI反应评分、推荐方案]

ICI Response Score: XX/100

ICI反应评分: XX/100

Response Likelihood: [HIGH/MODERATE/LOW] Confidence: [HIGH/MODERATE/LOW] Expected ORR: XX-XX%

反应可能性: [高/中/低] 置信度: [高/中/低] 预期ORR: XX-XX%

Score Breakdown

评分拆解

Component	Value	Score	Max
TMB	XX mut/Mb	XX	30
MSI Status	MSI-H/MSS	XX	25
PD-L1	XX%	XX	20
Neoantigen Load	XX est.	XX	15
Sensitivity Bonus	+XX	XX	10
Resistance Penalty	-XX	XX	-20
TOTAL		XX	100

成分	数值	得分	满分
TMB	XX mut/Mb	XX	30
MSI状态	MSI-H/MSS	XX	25
PD-L1	XX%	XX	20
新抗原负荷	XX（估算）	XX	15
敏感突变加分	+XX	XX	10
耐药突变扣分	-XX	XX	-20
总计		XX	100

Patient Profile

患者特征

Cancer Type: [cancer]
Mutations: [list]
TMB: XX mut/Mb [classification]
MSI Status: [MSI-H/MSS/Unknown]
PD-L1: XX% [scoring method]

癌症类型: [癌症名称]
突变: [列表]
TMB: XX mut/Mb [分类]
MSI状态: [MSI-H/MSS/未知]
PD-L1: XX% [评分方法]

Biomarker Analysis

生物标志物分析

TMB Analysis

TMB分析

[TMB classification, cancer-specific context, FDA TMB-H status]

[TMB分类、癌症特异性背景、FDA TMB-H状态]

MSI/MMR Status

MSI/MMR状态

[MSI status, MMR gene mutations, FDA MSI-H approvals]

[MSI状态、MMR基因突变情况、FDA MSI-H获批情况]

PD-L1 Expression

PD-L1表达

[PD-L1 level, cancer-specific thresholds, scoring method]

[PD-L1水平、癌症特异性阈值、评分方法]

Neoantigen Burden

新抗原负荷

[Estimated neoantigen count, quality assessment, mutation types]

[估算新抗原数量、质量评估、突变类型]

Mutation Analysis

突变分析

Driver Mutations

驱动突变

[Analysis of each mutation - oncogenic role, ICI implications]

[每个突变的分析：致癌作用、ICI相关性]

Resistance Mutations

耐药突变

[Any STK11, PTEN, JAK1/2, B2M, KEAP1 etc. with penalties]

[任何STK11、PTEN、JAK1/2、B2M、KEAP1等突变及扣分]

Sensitivity Mutations

敏感突变

[Any POLE, PBRM1, DDR genes with bonuses]

[任何POLE、PBRM1、DDR基因等突变及加分]

Immune Microenvironment

免疫微环境

[Hot/cold classification, immune gene expression data]

[热/冷分类、免疫基因表达数据]

ICI Drug Recommendation

ICI药物推荐

Primary Recommendation

首选推荐

[Drug name] - [monotherapy/combination]

Evidence: [FDA approval, trial data]
Expected response: XX-XX%
Key trial: [trial name/NCT#]

[药物名称] - [单药/联合]

证据: [FDA获批、临床试验数据]
预期反应: XX-XX%
关键试验: [试验名称/NCT编号]

Alternative Options

替代选项

[Alternative 1] - [rationale]
[Alternative 2] - [rationale]

[替代方案1] - [理由]
[替代方案2] - [理由]

Combination Strategies

联合策略

[ICI+ICI, ICI+chemo, ICI+targeted recommendations]

[ICI+ICI、ICI+化疗、ICI+靶向治疗推荐]

Clinical Evidence

临床证据

[Key trials, response rates, PFS/OS data for this cancer + biomarker profile]

[针对该癌症+生物标志物特征的关键试验、反应率、PFS/OS数据]

Resistance Risk

耐药风险

Risk Level: [LOW/MODERATE/HIGH]
Key Factors: [list resistance mutations/mechanisms]
Mitigation: [combination strategies]

风险等级: [低/中/高]
关键因素: [耐药突变/机制列表]
缓解策略: [联合治疗策略]

Monitoring Plan

监测方案

Response assessment: [schedule]
Biomarkers to track: [ctDNA, imaging, labs]
irAE monitoring: [schedule]
Resistance monitoring: [when to suspect progression]

反应评估: [时间表]
需追踪的生物标志物: [ctDNA、影像、实验室检查]
irAE监测: [时间表]
耐药监测: [何时怀疑进展]

Alternative Strategies (if ICI unlikely effective)

替代策略（若ICI可能无效）

[Targeted therapy, chemotherapy, clinical trials]

[靶向治疗、化疗、临床试验]

Evidence Grading

证据分级

Finding	Evidence Tier	Source
[finding 1]	T1 (FDA/Guidelines)	[source]
[finding 2]	T2 (Clinical trial)	[source]

结论	证据层级	来源
[结论1]	T1（FDA/指南）	[来源]
[结论2]	T2（临床试验）	[来源]

Data Completeness

数据完整性

Biomarker	Status	Impact
TMB	Provided/Estimated/Unknown	XX points
MSI	Provided/Unknown	XX points
PD-L1	Provided/Unknown	XX points
Neoantigen	Estimated	XX points
Mutations	X provided	+/-XX points

生物标志物	状态	影响
TMB	提供/估算/未知	XX分
MSI	提供/未知	XX分
PD-L1	提供/未知	XX分
新抗原	估算	XX分
突变	提供X个	+/-XX分

Missing Data Recommendations

缺失数据推荐

[What additional tests would improve prediction accuracy]

Generated by ToolUniverse Immunotherapy Response Prediction Skill Sources: OpenTargets, CIViC, FDA, DrugBank, PubMed, IEDB, HPA, cBioPortal

---

[哪些额外检测可提高预测准确性]

由ToolUniverse免疫治疗反应预测Skill生成 来源: OpenTargets、CIViC、FDA、DrugBank、PubMed、IEDB、HPA、cBioPortal

---

Evidence Tiers

证据层级

Tier	Description	Source Examples
T1	FDA-approved biomarker/indication	FDA labels, NCCN guidelines
T2	Phase 2-3 clinical trial evidence	Published trial data, PubMed
T3	Preclinical/computational evidence	Pathway analysis, in vitro data
T4	Expert opinion/case reports	Case series, reviews

层级	描述	来源示例
T1	FDA获批生物标志物/适应症	FDA标签、NCCN指南
T2	2-3期临床试验证据	已发表试验数据、PubMed
T3	临床前/计算证据	通路分析、体外数据
T4	专家意见/病例报告	病例系列、综述

Use Case Examples

用例示例

Use Case 1: NSCLC with High TMB

用例1：高TMB NSCLC

Input: "NSCLC, TMB 25, PD-L1 80%, no STK11 mutation" Expected: ICI Score 70-85, HIGH response, pembrolizumab monotherapy recommended

输入: "NSCLC, TMB 25, PD-L1 80%, no STK11 mutation" 预期: ICI评分70-85，高反应，推荐pembrolizumab单药

Use Case 2: Melanoma with BRAF

用例2：BRAF突变黑色素瘤

Input: "Melanoma, BRAF V600E, TMB 15, PD-L1 50%" Expected: ICI Score 50-65, MODERATE response, discuss ICI vs BRAF-targeted

输入: "Melanoma, BRAF V600E, TMB 15, PD-L1 50%" 预期: ICI评分50-65，中反应，讨论ICI vs BRAF靶向治疗

Use Case 3: MSI-H Colorectal

用例3：MSI-H结直肠癌

Input: "Colorectal cancer, MSI-high, TMB 40" Expected: ICI Score 80-95, HIGH response, pembrolizumab first-line

输入: "Colorectal cancer, MSI-high, TMB 40" 预期: ICI评分80-95，高反应，推荐pembrolizumab一线治疗

Use Case 4: Low Biomarker NSCLC

用例4：低生物标志物NSCLC

Input: "NSCLC, TMB 2, PD-L1 <1%, STK11 mutation" Expected: ICI Score 5-20, LOW response, chemotherapy preferred

输入: "NSCLC, TMB 2, PD-L1 <1%, STK11 mutation" 预期: ICI评分5-20，低反应，推荐化疗

Use Case 5: Bladder Cancer

用例5：膀胱癌

Input: "Bladder cancer, TMB 12, PD-L1 10%, no resistance mutations" Expected: ICI Score 45-55, MODERATE response, ICI+chemo or maintenance

输入: "Bladder cancer, TMB 12, PD-L1 10%, no resistance mutations" 预期: ICI评分45-55，中反应，推荐ICI+化疗或维持治疗

Use Case 6: Checkpoint Inhibitor Selection

用例6：检查点抑制剂选择

Input: "Which ICI for NSCLC with PD-L1 90%?" Expected: Pembrolizumab monotherapy first-line, evidence from KEYNOTE-024

输入: "Which ICI for NSCLC with PD-L1 90%?" 预期: 推荐pembrolizumab单药一线治疗，证据来自KEYNOTE-024试验