5月week1 文献阅读：TheCancerCell Line Encyclopedia enables predictive modelling of anticancer drug sensitivity

癌症细胞系百科全谱确立模型预测抗癌药物敏感性

Abstract

• The systematic translation of cancer genomic data into knowledge of tumour biology and therapeutic possibilities remains challenging.

  系统地将癌症基因组数据转换为肿瘤生物学知识和治疗可能性仍然具有挑战性。

• Such efforts should be greatly aided by robust preclinical model systems that reflect the genomic diversity of human cancers and for which detailed genetic and pharmacological annotation is available.

  这些努力应得到强有力的临床前模型系统的大力支持，这些模型系统反映了人类癌症的基因组多样性，并具有详细的遗传和药理注释。

• Here we describe the Cancer Cell Line Encyclopedia (CCLE): a compilation of gene expression, chromosomal copy number and massively parallel sequencing data from 947 human cancer cell lines.

  在这里，我们描述了癌症细胞系百科全书(CCLE):从947个人类癌症细胞系汇编基因表达、染色体拷贝数和大规模并行测序数据。

• When coupled with pharmacological profiles for 24 anticancer drugs across 479 of the cell lines, this collection allowed identification of genetic, lineage, and gene-expression-based predictors of drug sensitivity.

  当横跨其中479个细胞系与24种抗癌药物的药理学特性相结合时，该集合收集识别遗传、谱系和基于基因表达的药物敏感性预测因子。

• In addition to known predictors, we found that plasma cell lineage correlated with sensitivity to IGF1 receptor inhibitors;

  除了已知的预测因子，我们还发现浆细胞谱系与IGF1受体抑制剂的敏感性相关;

• AHR expression was associated with MEK inhibitor efficacy in NRAS-mutant lines;

  AHR表达与MEK抑制剂在nras突变株中的作用有关;

• and SLFN11 expression predicted sensitivity to topoisomerase inhibitors.

  SLFN11表达预测对拓扑异构酶抑制剂的敏感性。

• Together, our results indicate that large, annotated cell-line collections may help to enable preclinical stratification schemata for anticancer agents.

  总之，我们的结果表明，大规模、注释的细胞系收集可能有助于为抗癌药物提供临床前分层模式。

• The generation of genetic predictions of drug response in the preclinical setting and their incorporation into cancer clinical trial design could speed the emergence of ‘personalized’ therapeutic regimens.

  在临床前环境中产生药物反应的遗传预测，并将其纳入癌症临床试验设计中，可能会加快“个性化”治疗方案的出现.

First paragraph

• Human cancer cell lines represent a mainstay of tumour biology and drug discovery through facile experimental manipulation, global and detailed mechanistic studies, and various high-throughput applications.

  人类癌细胞系是肿瘤生物学和药物发现的支柱，通过简单的实验操作，全球和详细的机制研究，以及各种高通量的应用。

• Numerous studies have used cell-line panels annotated with both genetic and pharmacological data, either within a tumour lineage or across multiple cancer types.

  许多研究都使用了带有遗传和药理数据注解的细胞系面板，这些数据要么来自肿瘤谱系，要么来自多种癌症类型。

• Although affirming the promise of systematic cell line studies, many previous efforts were limited in their depth of genetic characterization and pharmacological interrogation.

  虽然肯定了系统细胞系研究的前景，但以前的许多努力在其遗传特性和药理学研究的深度上受到限制。

(目前系统细胞系研究的现状)

Second

• To address these challenges, we generated a large-scale genomic data set for 947 human cancer cell lines, together with pharmacological profiling of 24 compounds across ,500 of these lines.

  为了应对这些挑战，我们为947个人类癌细胞系生成了大规模基因组数据集，并对24种化合物在其中500个细胞系中的进行了药理学分析。

• The resulting collection, which we termed the Cancer Cell Line Encyclopedia (CCLE), encompasses 36 tumour types (Fig. 1a and Supplementary Table 1;see also http://www.broadinstitute.org/ccle).

  最终的收集，我们称之为癌症细胞系百科全书(CCLE)，包含36种肿瘤类型(图1a和补充表1;也参见http://www.broadinstitute.org/ccle)。
  

  fig1a
  
  
  tableS1-1

tableS1-2

tableS1-3

• Supplementary Table 1: CCLE cell lines and associated data.

  补充表1:CCLE细胞系及其相关数据。

• The “CCLE name” consists of alpha-numeric characters from the Primary name and the site of the primary tumor from which the cell line was derived.

  “CCLE名称”由来自主名称的字母数字字符和源自该细胞系的原发肿瘤的位置组成。

• The “Notes” field describes cell lines that appear to derive from the same individual based on SNP fingerprint matching.

  “Notes”字段描述了基于SNP指纹匹配的来自同一个体的细胞系。

• “Expression arrays”, “SNP arrays”, “OncoMap”, “Hybrid capture/sequencing” and “Drug sensitivity profiling” refer to the existence of these data to date for the specified cell line.

  “表达阵列”、“SNP阵列”、“OncoMap”、“混合捕获/测序”和“药物敏感性分析”是指到目前为止这些数据存在于指定的细胞系中。

• All cell lines were characterized by several genomic technology platforms.

  所有的细胞系都由多个基因组技术平台进行了鉴定。

• The mutational status of >1,600 genes was determined by targeted massively parallel sequencing, followed by removal of variants likely to be germline events (Supplementary Methods).

  多于1600个基因的突变状态是通过靶向大规模并行测序确定的，然后去除可能是种系事件的变异(补充方法)。

• Moreover,Moreover, 392 recurrent mutations affecting 33 known cancer genes were assessed by mass spectrometric genotyping (Supplementary Table 2 and Supplementary Fig. 1). DNA copy number was measured using high-density single nucleotide polymorphism arrays (Affymetrix SNP 6.0;Supplementary Methods).

  此外，采用质谱法基因分型(补充表2和补充图1)评估了影响33个已知已知癌症基因的392例复发性突变(补充方法为Affymetrix SNP 6.0)。

  
  
  tableS2-1
  
  
  tableS2-2
  
  
  tableS2-3
  
  
  tableS2-4

tableS2-5

tableS2-6

figs1

• Supplementary Table 2: OncoMap assays and associated mutations.

  补充表2:OncoMap检测和相关突变。

• Details of the 456 mass spectrometric assays that were used to assess the presence 392 mutations in 33 genes are shown.

  本文详细介绍了用于评估33个基因中392个突变的456种质谱分析方法。

• The genomic positions refer to hg18 genomic coordinates.

  基因组位置参照hg18基因组坐标。

• Finally,messenger RNA expression levels were obtained for each of the lines using Affymetrix U133 plus 2.0 arrays.

  最后，使用Affymetrix U133 + 2.0阵列获得每条直线的mrna表达水平。

• These data were also used to confirm cell line identities (Supplementary Methods and Supplementary Figs 2–4)

  这些数据也被用来确认细胞系的身份(补充方法和补充图2-4)

  
  
  Supplementary Figure 2

Supplementary Figure 2: Venn diagram of overlap between the cell lines in the CCLE, the Sanger Institute’s Cancer Cell Line Project (http://www.sanger.ac.uk/genetics/CGP/CellLines/) and the GlaxoSmithKline (GSK) Cancer Cell Line Genomic Profiling dataset (https://cabig.nci.nih.gov/caArray_GSKdata/).

补充图2:CCLE细胞系、Sanger研究所癌症细胞系项目(http://www.sanger.ac.uk/genetics/CGP/CellLines/)和GlaxoSmithKline (GSK)癌症细胞系Genomic Profiling dataset (https://cabig.nci.nih.gov/caArray_GSKdata/)之间的重叠。

Supplementary Figure 3

Supplementary Figure 3: Distribution of the percentages of SNP identity across cell lines.

补充图3:SNP标识在细胞系间的百分比分布。

• The boxplot (orange) corresponds to the same values shown as a histogram in blue.

  箱线图(橙色)对应的值与用蓝色表示的直方图相同。

• Values above 80% identity denote cell line pairs presumed to derive from the same individual

  80%以上的值表示假定来自同一个体的细胞系对

  
  
  Supplementary Figure 4

Supplementary Figure 4: Principal component analysis of expression data from cell lines and primary tumors, for the 1,000 most varying genes.

补充图4:对1000个差异最大的基因的细胞系和原发肿瘤的表达数据进行主成分分析。

• Those derived from solid tumors are shown in orange, and those corresponding to hematopoietic lineages are shown in blue.

  来自实体肿瘤的用橙色表示，与造血谱系相对应的用蓝色表示。

• This figure indicates that there was no cross-contamination between hematopoietic and solid tumor-derived cell lines.

  这一图表明造血细胞系和实体肿瘤来源细胞系之间没有交叉污染。

• Expression data was also used to ascertain the original lineage of certain cell lines.

  表达数据也被用来确定某些细胞系的原始谱系。

(500细胞系基因表达测序，33个基因中392个突变的456种质谱分析方法，使用Affymetrix U133 + 2.0阵列获得每条直线的mrna表达水平确认细胞系的身份，可利用测序结果进行突变检测和区别细胞系)

Third

• We next measured the genomic similarities by lineage between CCLE lines and primary tumours from Tumorscape,expO,MILE and COSMIC data sets (Fig. 1b–d and Supplementary Methods).

  接下来，我们在Tumorscape、expO、MILE和COSMIC数据集(图1b-d和补充方法)中，通过谱系来测量CCLE和原发性肿瘤之间的基因组的相似性

  
  
  fig1b

fig1c

fig1d

• For most lineages, a strong positive correlation was observed in both chromosomal copy number and gene expression patterns (median correlation coefficients of0.77, range 0.52–0.94, P<10-15, for copy number, and 0.60, range50.29–0.77, P,10215, for expression, respectively;Fig. 1b, c and Supplementary Tables 3 and 4), as has been described previously.

  大多数谱系的染色体拷贝数与基因表达模式均存在较强的正相关关系(拷贝数的中位相关系数为0.77，范围0.52 - 0.94,P<10-15;表达量的中位相关系数为0.60，范围0.29 - 0.77,P<10-15);图1b, c和补充表3和4)，如前所述。

• A positive correlation was also observed for point mutation frequencies (median correlation coefficient 0.71, range 0.06– 0.97, P,1022 for all but 3 lineages;Supplementary Fig. 5), even when TP53 was removed from the data set (median correlation coefficient5 0.64, range520.31–0.97, P,1022 for all but 3 lineages;Fig. 1d and Supplementary Table 5).

  点突变频率(中位相关系数0.71，范围0.06 - 0.97,P<10-22，除3个谱系外，其余均为正相关;补充图5)，即使剔除TP53(中位相关系数 0.64，范围0.31 - 0.97,P<10-22，除3个谱系外，其余均为正相关;Fig1d和补充表5)

• Thus, with relatively few exceptions (Sup- plementary Information), the CCLE may provide representative genetic proxies for primary tumours in many cancer types.

  因此，除了相对较少的例外(补充信息)，CCLE可能为许多癌症类型的原发性肿瘤提供有代表性的基因组信息。

  (在Tumorscape、expO、MILE和COSMIC数据集(图1b-d和补充方法)中，通过谱系来测量CCLE和原发性肿瘤之间的基因组的相似性，说明CCLE可能为许多癌症类型的原发性肿瘤提供有代表性的基因组信息)

Fourth paragraph

• Given the pressing clinical need for robust molecular correlates of anticancer drug response, we incorporated a systematic framework to ascertain molecular correlates of pharmacological sensitivity in vitro.

  鉴于临床迫切需要强有力的抗癌药物反应的分子相关性，我们纳入了一个系统的框架，以确定体外药物敏感性的分子相关性。

• First, 8-point dose–response curves for 24 compounds (targeted and cytotoxic agents) across 479 cell lines were generated (Supplementary Tables 1 and 6, and Supplementary Methods).

  首先，生成了479个细胞系中24种化合物(靶向和细胞毒性药物)的8点剂量-反应曲线(补充表1和6，以及补充方法)。

• These curves were represented by a logistical sigmoidal function with a maximal effect level (Amax), the concentration at half-maximal activity of the compound (EC50), a Hill coefficient representing the sigmoidal transition, and the concentration at which the drug response reached an absolute inhibition of 50% (IC50)

  这些曲线由一个具有最大作用水平(Amax)的物流s型函数表示，化合物在半最大活性时的浓度(EC50)，表示s型转变的Hill系数，以及药物反应达到50%绝对抑制的浓度(IC50)

  (药物分析和基因表达联系的8点剂量反应曲线函数意义)

Fivth paragraph

• Broadly active compounds, exemplified by the HDAC inhibitor LBH589 (panobinostat), showed a roughly even distribution of Amax and EC50 values across most cell lines (Fig. 2a).

  以HDAC抑制剂LBH589 (panobinostat)为例的广泛活性化合物显示，Amax和EC50值在大多数细胞系中分布大致均匀(图2a)。

• In contrast, the RAF inhibitor PLX4720 had a more selective profile: Amax or EC50 values for most cell lines could be categorized as ‘sensitive’ or ‘insensitive’ to PLX4720, with sensitive lines enriched for the BRAFV600E mutation (Fig. 2a).

  相比之下，RAF抑制剂PLX4720具有更强的选择性:大多数细胞系的Amax或EC50值可以被归类为对PLX4720“敏感”或“不敏感”，BRAFV600E突变丰富了敏感系(图2a)。

  
  
  fig2a

• To capture simultaneously the efficacy and potency of a drug, we designated an ‘activity area’ (Fig. 2b and Supplementary Fig. 6). The 24 compounds profiled showed wide variations in activity area, and those with similar mechanisms of action clustered together (Supplementary Fig. 7)

  为了同时捕获药物的有效性和效力，我们指定了一个“活性区”(图2b和补充图6)。所描述的24种化合物的活性区差异很大，并且具有相似作用机制的化合物聚集在一起(补充图7)。

  
  
  fig2b

(药物的有效性和有效力“活力区的指定”，例细胞系的Amax 值和EC50值可以被归类为“敏感”或“不敏感”，24种药物的活力区差异很大。)

Sixth paragraph

• Genomic correlates of drug sensitivity maybe extractedbypredictive models using machine learning techniques.

  药物敏感性的基因组关联可能通过使用机器学习技术的预测模型提取出来。

• We therefore assembled all CCLE genomic data types into a matrix wherein each feature was converted to a z-score across all lines (Supplementary Methods).

  因此，我们将所有CCLE基因组数据类型组装成一个矩阵，其中每个特征在所有行上转换为z分数(补充方法)。

• Next, we adapted a categorical modelling approach that used a naive Bayes classification and discrete sensitivity calls, or an elastic net regression analysis for continuous sensitivity measurements.

  接下来，我们采用了一种分类建模方法，该方法使用朴素贝叶斯分类和离散灵敏度调用，或者对连续灵敏度测量使用弹性净回归分析。

• Both approaches were applied to all compounds and genomic data with or without gene expression features (Supplementary Methods).

  两种方法均适用于所有具有或不具有基因表达特征的化合物和基因组数据(补充方法)。

• Prediction performance was determined using tenfold cross-validation, and the elastic net features were bootstrapped to retain only those that were consistent across runs (Supplementary Methods).

  使用十倍交叉验证确定预测性能，并引导弹性网络特性，仅保留那些跨运行一致的特性(补充方法)。

  （利用统计学习技术，对药物敏感性判断建立判断模型（朴素贝叶斯分类和离散灵敏））

Seventh paragraph

• 50,000 input features, the regression-based analysis identified multiple known features as top predictors ofsensitivity to several agents (Supplementary Table 7 and Supplementary Figs 8 and 9), with robust cross-validated performance (Supplementary Fig. 10 and 11).

  5万个输入特征，基于回归分析确定了多个已知特征作为对多个agent敏感性的顶级预测因子(补充表7和补充图8和9)，具有健壮的交叉验证性能(补充图10和11)。

• For example, activating mutations in BRAFandNRAS were among the top four predictors of sensitivity in models generated for the MEK inhibitor PD-0325901 (ref. 10) (Fig. 2c).

  例如，在MEK抑制剂PD-0325901生成的模型中，激活BRAFandNRAS突变是敏感性的四大预测因子之一(参考文献10)(图2c)。

• Additional predictive features for MEK inhibition included expression ofPTEN, PTPN5 and SPRY2 (which encodes a regulator of MAPK output).

  MEK抑制的其他预测特征包括pten、PTPN5和SPRY2(编码mapk输出的调节器)的表达。

• KRAS mutations were also identified, albeit with a lower predictive value (Fig. 2c, Supplementary Tables 8 and 9 and Supplementary Fig. 8)

  KRAS突变也被发现，尽管预测值较低(图2c，补充表8和9和补充图8)。

  
  
  fig2c

（利用基因表达突变预测药物敏感性的模型）

Eighth paragraph

• Other top predictors included EGFR mutations and ERBB2 amplification/overexpression for erlotinib and lapatinib , respectively;BRAFV600E for RAF inhibitors (PLX4720 (ref. 18) and RAF265);

  其他预测因子包括EGFR突变和ERBB2分别在erlotinib和lapatinib上的扩增/过表达;BRAFV600E在RAF抑制剂上的表达(PLX4720(参考文献18)和RAF265);

• HGF expression and MET amplification for the MET/ALK inhibitor PF- 2341066 (ref. 19);

  MET/ALK抑制剂PF- 2341066的HGF表达及MET扩增(参考文献19);

• and MDM2 overexpression for Nutlin-3 (ref. 20) sensitivity.

  MDM2过表达为Nutlin-3(文献20)敏感性。

• Variants affecting the EXT2 gene, which encodes a glycosyltransferase involved in heparin sulphate biosynthesis, were signifi- cantly correlated with erlotinib effects (Supplementary Fig. 12).

  影响EXT2基因(编码参与硫酸肝素生物合成的糖基转移酶)的变异与厄洛替尼效应显著相关(补充图12)。

• This observation is intriguing in light of a report linking heparin sulphate with erlotinib sensitivity.

  鉴于一份报告将硫酸肝素与厄洛替尼敏感性联系起来，这一观察结果很有趣。

• In addition, NQO1 expression was identified as the top predictive feature for sensitivity to the Hsp90 inhibitor 17AAG, a quinone moiety metabolized by NAD(P)H:quinone oxido- reductase (NQO1).

此外，NQO1的表达被认为是Hsp90抑制剂17AAG敏感性的首要预测特征。17AAG是由NAD(P)H代谢的醌类分子:醌氧化还原酶(NQO1)。

• NQO1 produces a high-potency intermediate (17AAGH2), and has previously been identified as a potential biomarker for Hsp90 inhibitors.

  NQO1产生高效中间体(17AAGH2)，并已被确定为潜在的Hsp90抑制剂的生物标志物。

  （相关药物敏感性的预测因子，某些基因突变的表达被确立为药物敏感性的生物标志物，并有相关文献论证）

Nineth paragraph

• Because some genetic/molecular alterations occur commonly in specific tumour types, lineage may become a confounding factor in predictive analyses.

  由于某些遗传/分子改变通常发生在特定的肿瘤类型中，谱系可能成为预测分析中的混杂因素。

• Indeed, a classifier built using the entire cell-line data set performed suboptimally when applied exclusively to melanoma-derived cell lines (Fig. 2d), whereas a model built with only melanoma cell lines performed better (Fig. 2d).

  事实上，当仅应用于黑色素瘤衍生的细胞系时，使用整个细胞系数据集构建的分类器的性能较差(图2d)，而仅使用黑色素瘤细胞系构建的模型性能较好(图2d)。

  
  
  fig2d

• Predictive features in the melanoma-only model showed a strong overexpression of genes regulated by the transcription factors MITF and SOX10 (Supplementary Table 10), which may also help predict RAF inhibitor drug sensitivity in melanoma cell lines.

  单纯黑色素瘤模型的预测特征显示，受转录因子MITF和SOX10调控的基因具有较强的过表达(补充表10)，这也可能有助于预测RAF抑制剂对黑色素瘤细胞株的药物敏感性。

  (可利用整个细胞系数据集构建的分类器，也可用仅使用某种细胞系构建分类器，（如黑色素瘤细胞系），根据使用预测敏感的功能，各系谱数据集构成的分类器的性能也不同)

Tenth paragraph

• Nonetheless, lineage emerged as the predominant predictive feature for several compounds.

  尽管如此，谱系还是成为几种化合物的主要预测特征。

• For example, elastic net studies of the HDAC inhibitor panobinostat identified haematological lineages as predictors of sensitivity (Fig. 2e and Supplementary Fig. 9). Interestingly, most clinical responses to panobinostat and related compounds (for example, vorinostat and romidepsin) have been observed in haematological cancers.

  例如，HDAC抑制剂panobinostat的弹性网络研究将血流变学谱系确定为敏感性的预测因子(图2e和补充图9)。

• Similarly, most multiple myeloma cell lines (12 of 14 lines tested) exhibited enhanced sensitivity to the IGF1 receptor inhibitor AEW541 (Fig. 2f and Supplementary Figs 8 and 9) and showed high IGF1 expression (Fig. 2f).

  同样，大多数多发性骨髓瘤细胞系(14个被测细胞系中的12个)对IGF1受体抑制剂AEW541表现出更高的敏感性(图2f和补充图8和9)，并表现出高IGF1表达(图2f)。

  
  
  fig2ef

• Interestingly, elevated IGF1R expression also correlated with AEW541 sensitivity(SupplementaryFig.9). The CCLE results indicate that multiple myeloma may be a promising indication for clinical trials of IGF1 receptor inhibitors and that these drugs may have enhanced efficacy in cancers with high IGF1 or IGF1R expression.

  有趣的是，IGF1R表达升高也与 AEW541敏感性相关(补充图9). CCLE结果表明，多发性骨髓瘤可能是临床试验IGF1 受体抑制剂的一个有希望的适应症，这些药物可能对IGF1或IGF1R高表达的癌症具有增强疗效的作用。

  （IGFIR表达升高和AEW541敏感性相关）

11

• Whereas BRAF and NRAS mutations are known single-gene predictors of sensitivity to MEK inhibitors, several ‘sensitive’ cell lines lacked mutations in these genes, whereas other lines harbouring these mutations were nonetheless ‘insensitive’ (Fig. 2c).

  虽然BRAF和NRAS突变是已知的对MEK抑制剂敏感的单基因预测因子，但这些基因中有几个“敏感”细胞系缺乏突变，而包含这些突变的其他细胞系却“不敏感”(图2c)。

• The elastic net regression model derived from the subset of cell lines with validated NRAS mutations identified elevated expression of the AHR gene (which encodes the aryl hydrocarbon receptor) as strongly correlated with sensitivity to the MEK inhibitor PD-0325901 (Fig. 3a).

  从已验证NRAS突变的细胞系子集中得到的弹性网络回归模型发现，AHR基因(编码芳基烃受体)的表达升高与MEK抑制剂PD-0325901的敏感性密切相关(图3a)。

  
  
  fig3a

• This find- ing was interesting in light of previous studies indicating that a related MEK inhibitor (PD-98059) may also function as a direct AHR antagonist.

  这一发现很有趣，因为先前的研究表明，一种相关的MEK抑制剂(PD-98059)也可能作为一种直接的AHR拮抗剂。

• We therefore hypothesized that the enhanced sensitivity of some NRAS-mutant cell lines to MEK inhibitors might relate to a coexistent dependence on AHR function

  因此，我们推测某些 NRAS突变细胞系对MEK抑制剂的敏感性增强可能与AHR功能的共存依赖有关.

  （利用NRAS突变的细胞系子集中得到的弹性网络回归模型发现，AHR基因(编码芳基烃受体)的表达升高与MEK抑制剂PD-0325901的敏感性密切相关，得到推测：某些 NRAS突变细胞系对MEK抑制剂的敏感性增强可能与AHR功能的共存依赖有关）

12

• To test this hypothesis, we first confirmed the correlation between AHR expression and sensitivity to MEK inhibitors in a subset of NRAS-mutant cell lines (Fig. 3b and Supplementary Fig. 13).

  为了验证这一假设，我们首先在nras突变细胞系的一个子集中证实了AHR表达与MEK抑制剂敏感性之间的相关性(图3b和补充图13)。

  
  
  fig3b

• Next, we performed short hairpin RNA (shRNA) knockdown of AHR in cell lines with high or low AHR expression (Fig. 3c).

  接下来，我们在AHR表达量高或低的细胞系中进行对AHR的短RNA (shRNA)敲除(图3c)。

  
  
  fig3c

• Silencing of AHR suppressed the growth of three NRAS-mutant cell lines with elevated AHR expression (Fig. 3d–f), but had no effect on the growth of two lines with low AHR expression (Fig. 3g, h). The growth inhibitory effect was confirmed with two additional shRNAs, where evidence for dose dependence was also apparent (Fig. 3i, j).

  有着促进AHR表达的短片段的AHR抑制三个NRAS-mutant细胞系的生长(图3 d-f),但较低的表达ANHR没有影响二个细胞系的增长(图3 g h)。生长抑制作用被证实有两个额外的成分shRNAs,在剂量依赖的证据也明显(图3 i, j)。

  
  
  figIj

• i, Left: proliferation of IPC-298 cells (high AHR) after introduction of additional shRNAs against AHR (shAHR_1 and shAHR_4;green and purple lines, respectively) or luciferase (control shLuc;blue line).

  i，左:添加针对AHR的shrna (shAHR_1和shAHR_4)后，IPC-298细胞的增殖(高AHR);

  绿色和紫色线，分别)或荧光素酶(对照shLuc;蓝色线)。

• Right: corresponding immunoblot analysis of AHR protein j, Equivalent studies as in i using SK-MEL-2 cells (high AHR).

  右图:AHR蛋白j的相应免疫印迹分析，与i中使用SK-MEL-2细胞(高AHR)进行等效研究。

• We also tested the hypothesis that allosteric MEK inhibitors may suppress AHR function by measuring the effect ofPD-0325901 and PD-98059 on endogenous CYP1A1 mRNA, a transcriptional target of AHR in some contexts.

  我们还通过测量pd -0325901和PD-98059对内源性CYP1A1 mRNA的影响，验证了变构MEK抑制剂可能抑制AHR功能的假说，CYP1A1 mRNA是AHR在某些情况下的转录靶点。

• Both compounds reduced CYP1A1 levels in NRAS-mutant melanoma cells (IPC-298 and SK-MEL-2;Fig.3k) but not in neuroblastoma cells (CHP-212;Fig.3k), indicating that other factors may govern CYP1A1 expression in the latter lineage.

  这两种化合物都降低了nras突变型黑色素瘤细胞(IPC-298和SK-MEL-2;图3k)中的CYP1A1水平，但在神经母细胞瘤细胞(cp -212;图3k)中没有降低，这表明在后者谱系中CYP1A1的表达可能受其他因素的控制。

  
  
  fig3k

• Together, these results suggest that AHR dependency may co-occur with MAP kinase activation in some NRAS-mutant cancer cells, and that elevated AHR may serve as a mechanistic biomarker for enhanced MEK inhibitor sensitivity in this setting.

  综上所述，这些结果表明，AHR依赖可能与MAP激酶活化在某些nras突变的癌细胞中同时发生，而升高的AHR可能作为在这种情况下增强MEK抑制剂敏感性的一种机制生物标志物。

  (验证假设的实验和分析过程)

  (高维回归如何帮助验证生物现象，产生新生物假说。

  ：

  假说过程：但这些基因中有几个“敏感”细胞系缺乏突变，而包含这些突变的其他细胞系却“不敏感”。

  从已验证NRAS突变的细胞系子集中得到的弹性网络回归模型发现，AHR基因(编码芳基烃受体)的表达升高与MEK抑制剂PD-0325901的敏感性密切相关。

  因此，我们推测某些 NRAS突变细胞系对MEK抑制剂的敏感性增强可能与AHR功能的共存依赖有关。

  实验：AHR的短RNA (shRNA)敲除，验证生物假说，

  结论：AHR依赖可能与MAP激酶活化在某些nras突变的癌细胞中同时发生，而升高的AHR可能作为在这种情况下增强MEK抑制剂敏感性的一种机制生物标志物)

13

• We also looked for markers predictive of response to several conventional chemotherapeutic agents (Supplementary Fig. 7 and Supplementary Table 6) and identified SLFN11 expression as the top correlate of sensitivity to irinotecan(Fig.4a),a camptothecin analogue that inhibits the topoisomerase I (TOP1)enzyme.

  我们还寻找了几种传统化疗药物反应的预测标志物(补充图7和补充表6)，并确定SLFN11的表达与对伊立替康(图4a)的敏感性相关(图4a)，这是一种抑制拓扑异构酶I (TOP1)的喜树碱类似物。

• SLFN11expression also emerged as the top predictor of topotecan sensitivity (another TOP1 inhibitor;Supplementary Figs 8 and 14).

  slfn11表达也成为topotecan敏感性(另一TOP1抑制剂;补充图8及14)。

• Overall, 12 of 16 lineages showed significant SLFN11 associations for topotecan or irinotecan sensitivity (Pearson’s r$0.2, Supplementary Fig. 14b).

  总的来说，16个谱系中有12个表现出与拓扑替康或伊立替康敏感性显著相关的SLFN11 (Pearson’s 0.2 r，补充图14b)。

• This finding was independently validated using data from the NCI-60 collection (Supplementary Fig. 15).

  使用NCI-60收集的数据独立验证了这一发现(补充图15)。

• SLFN11 knockdown did not affect steady-state growth sensitivity profiles (Supplementary Fig. 14d–f).

  SLFN11基因敲除不影响稳态生长敏感性曲线(补充图14d-f)。

  （传统化疗药物反应的预测标志物）

14

• All three Ewing’s sarcoma cell lines screened showed both high SLFN11 expression and sensitivity to irinotecan (Fig. 4b and Supplementary Fig. 14).

  所筛选的三个尤文氏肉瘤细胞株均表现出较高的SLFN11表达和对伊立替康的敏感性(图4b和补充图14)。

  
  
  fig4d

• Ewing’s sarcomas also exhibited the highest SLFN11 expression among 4,103 primary tumour samples spanning 39 lineages (Fig. 4c), suggesting that TOP1 inhibitors might offer an effective treatment option for this cancer type.

  Ewing的肉瘤在39个谱系的4103个原发肿瘤样本中也显示出最高的SLFN11表达(图4c)，这表明TOP1抑制剂可能为这种癌症类型提供一种有效的治疗选择。

• Towards this end, several ongoing trials in Ewing’s sarcoma are examining irinotecanbased combinations,or the addition of topotecan to standard regimens.

  为此，在尤文氏肉瘤中正在进行的几项试验正在研究以伊立替康为基础的联合疗法，或在标准治疗方案中添加托泊替康。

• For some lineages with high SLFN11 expression (for example, cervical adenocarcinoma),topoisomerase inhibitors already comprise a standard chemotherapy regimen.

  对于一些SLFN11高表达的谱系(例如，宫颈腺癌)，拓扑异构酶抑制剂已经包含一个标准的化疗方案。

• In other tumours where topoisomerase inhibitors are commonly used (for example, colorectal and ovarian cancers), a range of SLFN11 expression was observed, raising the possibility that high SLFN11 expression might enrich for tumours more likely to respond.

  在其他常用拓扑异构酶抑制剂的肿瘤中(例如，结直肠癌和卵巢癌)，观察到一系列SLFN11表达，这增加了高SLFN11表达可能丰富肿瘤更可能作出反应的可能性。

• If confirmed in correlative clinical studies, SLFN11 expression may offer a means to stratify patients for to poisomerase inhibitor treatment

  如果在相关的临床研究中得到证实，SLFN11的表达可能为对患者进行毒物酶抑制剂治疗的分层提供了一种手段。

  (利用所筛选的三个尤文氏肉瘤细胞株均表现出较高的SLFN11表达和对伊立替康的敏感性，类推对于一些SLFN11高表达的谱系(例如，宫颈腺癌)，拓扑异构酶抑制剂已经包含一个标准的化疗方案。在临床研究中进行证实)

15

• By assembling the CCLE, we have expanded the process of detailed annotation of preclinical human cancer models(http://www.broadinstitute.org/ccle).Genomic predictors of drug sensitivity revealed both known and novel candidate biomarkers of response.

  通过组装CCLE，我们扩展了临床前人类癌症模型的详细注释过程(http://www.broadinstitute.org/ccle)。药物敏感性的基因组预测因子揭示了已知的和新的反应候选生物标志物。

• Even within genetically defined sub-populations—orwhenagents were broadly active without clear genetic targets—elastic net modelling studies identified key predictors o rmechanistic effectors of drug response.

  即使在基因定义的亚种群中，或者当药物在没有明确基因靶点的情况下广泛活跃时，弹性网络模型研究也确定了药物反应的关键预测因子和机制效应因子。

• Additional efforts that increase the scale and provide complementary types of information(for example,whole-genome/transcriptome sequencing,epigenetic studies, metabolic profiling or proteomic/phosphoproteomic analysis) should enable additional insights.

  此外，提高伦理水平和提供基本类型信息能力(例如，全基因组/转录测序、表观遗传学研究、代谢侧写或蛋白质组/磷蛋白质组分析)应该能够提供额外的见解。

• In the future, comprehensive and tractable cell-linesystemsprovidedthroughthisandotherefforts27mayfacilitate numerous advances in cancer biology and drug discovery.

  在未来，通过这项和这项努力所提供的全面的、可处理的细胞系系统可能会促进癌症生物学和药物发现方面的许多进展。

  （CCLE弹性网络模型研究预测药物敏感性当前研究进展以及未来进展方向）

16 METHODS SUMMARY

• Mutationinformationwasobtained both by using massively parallel sequencing of .1,600 genes (Supplementary Table12)andbymassspectrometricgenotyping(OncoMap),whichinterrogated 492mutationsin33knownoncogenesandtumoursuppressors.

  突变信息是通过对. 1600个基因进行大规模并行测序(补充表12)和通过质谱基因分型(OncoMap)获得的。

• Genotyping/copy number analysis was performed using Affymetrix Genome-Wide Human SNP Array 6.0 and expression analysis using the GeneChip Human Genome U133 Plus2.0Array.

  基因分型/拷贝数分析采用Affymetrix全基因组SNP Array 6.0进行，表达分析采用GeneChip Human Genome U133 Plus2.0Array。

• Eight-pointdose–responsecurvesweregeneratedfor24anticancer drugs using an automated compound-screening platform.

  8点剂量反应安全系统使用一个自动化的化合物筛选平台生成了24种抗癌药物。

• Compound sensitivity data were used for two types of predictive models that used the naive Bayes classifier or the elastic net regression algorithm.

  采用朴素贝叶斯分类器和弹性网络回归算法对两类预测模型进行了复合敏感性数据的处理。

• The effects of AHR expression silencing on cell viability were assessed by stable expression of shRNA lentiviral vectors targetingeitherthis gene or luciferase as control.

  以shRNA慢病毒载体的稳定表达为对照，观察AHR表达沉默对细胞活力的影响。

• The effect of compound treatment on AHR target gene expression was assessed by quantitative RT–PCR.

  采用定量荧光定量pcr (pcr - pcr)技术检测复方制剂对肿瘤的治疗效果。

• A full description of the Methods is included in Supplementary Information

  补充资料中载有对这些方法的全面说明