Kaplan-Meierプロットに付加情報を追加するマクロの作成

Similar documents
Kaplan-Meierプロットに付加情報を追加するマクロの作成

Microsoft PowerPoint - 【魚住】 発表原稿【Web公開用】.ppt [互換モード]

untitled

スライド 1

Presentation Title Goes Here

食道がん化学放射線療法後のsalvage手術

SAS Enterprise Guideによるデータ解析入門

*1 * Wilcoxon 2 2 t t t t d t M t N t M t n t N t n t N t d t N t t at ri

7 / 70 8 / 70 SAS Enterprise Guide, Windows SAS, SAS,,,, SAS SAS Enterprise Guide SAS: SAS Enterprise Guide:, SAS SAS Enterprise Guide... 5 / 70 6 / 7

SAS Enterprise Guideによるデータ解析入門

ODS GRAPHICS ON; ODS GRAPHICS ON; PROC TTEST DATA=SASHELP.CLASS SIDE=2 DIST=NORMAL H0=58 PLOTS(ONLY SHOWH0)=(SUMMARY); VAR HEIGHT;

Slide 1

JMP V4 による生存時間分析

スライド 1



VENTANA ALK D5F3 Rabbit Monoclonal Antibody OptiView ALK D5F3

ProVisionaire Control V3.0セットアップガイド

ERBITUX_WEBセミナー表紙台_2015.ai

PHREG プロシジャにおける 共変量調整解析に関連したオプション機能 魚住龍史 1 * 矢田真城 2 浜田知久馬 3 1 京都大学大学院医学研究科医学統計生物情報学 2 エイツーヘルスケア株式会社 3 東京理科大学 Investigating fascinating aspects associa


日本消化器外科学会雑誌第29巻第9号



Microsoft PowerPoint - 【配布・WEB公開用】SAS発表資料.pptx

R Commanderを用いたデータ解析

untitled

DATA Sample1 /**/ INPUT Price /* */ DATALINES

日本消化器外科学会雑誌第30巻第3号

E3200 BV FOLFOX4 FOLFOX4 PFS % 0.42, 0.65 p< log-rank % FOLFOX4 8.6% FOLFOX % FOLFOX4 259/ % FOL

untitled


VOL.39 S-3


1 2 3

EBM Reviews Ovid Ovid EBM Reviews ACP Journal Club Cochrane Database of Systematic Revie

6 Tgif William Chia-Wei Chang tgif 3.0 pixmap URL Tgif 6.1: Tgif

日本製薬工業協会シンポジウム 生存時間解析の評価指標に関する最近の展開ー RMST (restricted mean survival time) を理解するー 2. RMST の定義と統計的推測 2018 年 6 月 13 日医薬品評価委員会データサイエンス部会タスクフォース 4 生存時間解析チー

ベバシズマブ併用 FOLFOX/FOLFIRI療法

1.eps


course pptx

特殊病態下感染症における抗菌薬治験の手引き作成委員会報告書(案)

スライド 1

1 I EViews View Proc Freeze

スライド 1

スライド 1

VOL. 34 S-2 CHEMOTH8RAPY 913

_02.indd

Release Notes for JMP book

こんにちは由美子です

Gynecologic Oncology Trial and Investigation Consortium GOTIC-002

要旨 : データステップ及び SGPLOT プロシジャにおける POLYGON/TEXT ステートメントを利用した SAS プログラムステップフローチャートを生成する SAS プログラムを紹介する キーワード :SGPLOT, フローチャート, 可視化 2

52-2.indb

Release Notes.book

: (EQS) /EQUATIONS V1 = 30*V F1 + E1; V2 = 25*V *F1 + E2; V3 = 16*V *F1 + E3; V4 = 10*V F2 + E4; V5 = 19*V99

TM-m30 詳細取扱説明書

RT300i/RT140x/RT105i 取扱説明書

Microsoft Word - Meta70_Preferences.doc

TM-m30 詳細取扱説明書

胃癌診療 -内視鏡治療と予防の最前線-

RT300/140/105シリーズ 取扱説明書

スケーリング理論とはなにか? - --尺度を変えて見えること--

TM-T88VI 詳細取扱説明書


3 4 2

第3回委員会(2月20日)資料NO

SAS Web XML * ** * ** Web Data Analysis with SAS Input and Output of XML Data and Application to Real Estate Valuation Map Junnosuke Matsushima*, Hiro

1_2.eps

8 The Bulletin of Meiji University of Integrative Medicine API II 61 ASO X 11 7 X-4 6 X m 5 X-2 4 X 3 9 X 11 7 API 0.84 ASO X 1 1 MR-angio

Transcription:

1 / 47 Kaplan-Meier A SAS macro for extended Kaplan-Meier plots 1, 2,3 1 2 3 Kengo Nagashima 1, Yasunori Sato 2,3 1 Department of Parmaceutical Technochemistry, Josai University 2 School of Medicine, Chiba University 3 Department of Biostatistics, Harvard School of Public Health SAS / & 2010 2010 7 27

2 / 47 %km_data() Kaplan-Meier Proportion of overall survival 1.0 0.8 0.6 0.4 0.2 0.0 MST 95% C.I. 1: high-risk 7.8 [4.4, 9.0) 2: middle-risk 11.3 [6.7,.) 3: low-risk. [8.2,.) No. at risk (1st entry: high, 2nd: middle, 3rd: low) 50 36 28 23 15 0 50 42 36 30 25 0 50 44 36 32 26 0 0.0 2.5 5.0 7.5 10.0 12.5 high-risk middle-risk low-risk Months after entry

( ) 1 (Kaplan-Meier ) 2 Log-rank ( ) 3 ( ). Kaplan-Meier.. 3 / 47

Kaplan-Meier Kaplan-Meier 1.0 Proportion of overall survival 0.8 0.6 0.4 0.2 0.0 high-risk middle-risk 0.0 2.5 5.0 7.5 10.0 12.5 Months after entry, Kaplan-Meier, log-rank P,, 4 / 47

(BMJ) No. at risk,, Pocock SJ, et al. BMJ 2008; 336: 1167. Fig. 4 5 / 47

(JAMA) No. at risk, P Regine WF, et al. JAMA 2008; 299(9): 1024. Fig. 2 6 / 47

(J Clin Oncol) No. at risk,,, P Reck M, et al. J Clin Oncol 2009; 27(8): 1231. Fig. 3 7 / 47

(Lancet) No. at risk,,, P Nordlinger B, et al. Lancet 2008; 371(9617): 1013. Fig. 2 8 / 47

(N Engl J Med) No. at risk, P Karapetis CS, et al. N Engl J Med 2008; 359(17): 1762. Fig. 1 9 / 47

10 / 47 %km_data()

11 / 47 %km_data() 1 Cox 2 3

%km_data() 1 LIFETEST Procedure / PHREG Procedure 2 3 GPLOT Procedure /* */ %include "&Path.kmdata_v213.sas"; /* */ %global color1 color2 scolor1 scolor2; %let color1 = cx445694; %let color2 = cxa23a2e; %let scolor1 = cxd4d9e8; %let scolor2 = cxf1cece; /* ( ) */ %km_data( D1, T, GroupC, Censor, 1, out = graph, anno = anno, CI = 1, censext = 1, Size = 2, atrisk = 1, atriskorder = 0 to 12.5 by 2.5, Step = 5, Label = "No. at risk (1st entry: high, 2nd: middle, 3rd: low)", Test = 1, TestX = 98, TestY = 97, Type = logrank, HR = 1, HRX = 98, HRY = 92 ); /* */ proc gplot data = Graph; plot (Sv1 Sv2 Sv3) * T / anno = anno overlay; run; quit; 12 / 47

13 / 47 1 data: 2 time: [data] 3 group: [data] 4 censor: [data] 5 censorv: [censor] 6 out: 7 anno: annotate ( annotate ) HP, 7

14 / 47 1.0 Proportion of overall survival 0.8 0.6 0.4 0.2 0.0 No. at risk (1st entry: high, 2nd: middle) 50 36 28 23 15 0 50 42 36 30 25 0 high-risk middle-risk 0.0 2.5 5.0 7.5 10.0 12.5 Months after entry

15 / 47 data D2; set D1; where Group in (1, 2); %km_data( D2, T, GroupC, Censor, 1, out = graph, anno = anno, censext = 1, cwidth = 20, Size = 1, Step = 5, afont = " Arial ", atrisk = 1, atriskorder = 0 to 12.5 by 2.5, Base = 0, Label = "No. at risk (1st entry: high, 2nd: middle)" ); atrisk = 1 atriskorder Base Label

16 / 47 1.0 Log-rank P = 0.067 Proportion of overall survival 0.8 0.6 0.4 0.2 0.0 high-risk middle-risk 0.0 2.5 5.0 7.5 10.0 12.5 Months after entry

17 / 47 data D2; set D1; where Group in (1, 2); %km_data( D2, T, GroupC, Censor, 1, out = graph, anno = anno, censext = 1, cwidth = 20, Size = 1, Step = 5, afont = " Arial ", Test = 1, TestX = 98, TestY = 97, Type = logrank ); Test = 1 TestX, TestY Type

18 / 47 Cox 1.0 2:middle-risk HR = 0.612 [0.360, 1.040] Proportion of overall survival 0.8 0.6 0.4 0.2 0.0 high-risk middle-risk 0.0 2.5 5.0 7.5 10.0 12.5 Months after entry

19 / 47 Cox data D2; set D1; where Group in (1, 2); %km_data( D2, T, GroupC, Censor, 1, out = graph, anno = anno, censext = 1, cwidth = 20, Size = 1, Step = 5, afont = " Arial ", HR = 1, HRX = 98, HRY = 97 ); HR = 1 HRX, HRY

20 / 47 Proportion of overall survival 1.0 0.8 0.6 0.4 0.2 1: high-risk 2: middle-risk MST 95% C.I. 7.8 [4.4, 9.0) 11.3 [6.7,.) 0.0 high-risk middle-risk 0.0 2.5 5.0 7.5 10.0 12.5 Months after entry

21 / 47 data D2; set D1; where Group in (1, 2); %km_data( D2, T, GroupC, Censor, 1, out = graph, anno = anno, censext = 1, cwidth = 20, Size = 1, Step = 5, afont = " Arial ", MST = 1, MlabX = 60, MmedX = 83, MciX = 98, MSTY = 95 ); MST = 1 MlabX, MmedX, MciX, MSTY

22 / 47 (1) 1.0 Proportion of overall survival 0.8 0.6 0.4 0.2 0.0 high-risk middle-risk 0.0 2.5 5.0 7.5 10.0 12.5 Months after entry

23 / 47 (1) data D2; set D1; where Group in (1, 2); %km_data( D2, T, GroupC, Censor, 1, out = graph, anno = anno, censext = 1, cwidth = 20, Size = 1, Step = 5, afont = " Arial " );, CI = 1

24 / 47 (2) 1.0 Proportion of overall survival 0.8 0.6 0.4 0.2 0.0 high-risk middle-risk 0.0 2.5 5.0 7.5 10.0 12.5 Months after entry

(2) data D2; set D1; where Group in (1, 2); %km_data( D2, T, GroupC, Censor, 1, out = graph, anno = anno, censext = 1, cwidth = 20, Size = 1, Step = 5, afont = " Arial " ); data Graph2; length vname $10.; set Graph; var=sv1; vname= Sv1 ; output; var=sv2; vname= Sv2 ; output; var=sl1; vname= zsl1 ; output; var=sl2; vname= zsl2 ; output; var=su1; vname= zsu1 ; output; var=su2; vname= zsu2 ; output; proc sort data = Graph2; by vname T; proc gplot data = Graph2; plot var * T = vname / legend = legend1; legend1 label = none position = (inside) mode = share across = 1 origin = (1, 0.2) value = (h = 1 "high-risk" "middle -risk") order = ("Sv1" "Sv2"); symbol1 i = steplj c="&color1." w=20; symbol2 i = steplj c="&color2." w=20; symbol3 i = steplj c="&scolor1." w=20 l=2; symbol4 i = steplj c="&scolor2." w=20 l=2; symbol5 i = steplj c="&scolor1." w=20 l=2; symbol6 i = steplj c="&scolor2." w=20 l=2; run; quit;, GPLOT Procedure Sv1, Sv2 (order ) 25 / 47

26 / 47 %km_data() %km_data( D2, T, GroupC, Censor, 1, out = graph, anno = anno, censext = 1, cwidth = 20 ); GPLOT Procedure, ( emf, eps )

27 / 47 ODS Graph LIFETEST Procedure ODS Graph SAS 9.2 ods listing gpath = "&Path." style = Statistical sge = on; ods graphics on / antialias = on border = off scale = on imagename = "Lifetest_ods" width = 6.33333333 in height = 4 in; proc lifetest data = D1 plots=(survival(atrisk=(0 to 12.5 by 2.5) test cl)); time T * Censor(1); strata GroupC; run; ods graphics off; ods listing close; ods listing; emf, eps

28 / 47 ODS Graph?

29 / 47 ODS Graph 1 ODS Graphics Editor 2 TEMPLATE Procedure

30 / 47 ODS Graphics Editor sge ( : ods listing sge = on;)

31 / 47 ODS Graphics Editor 1... TEMPLATE Procedure

32 / 47 TEMPLATE Procedure SAS ( )

33 / 47 ODS STYLE Statement style.statistical proc template; define style Styles.MyStatistical; parent = styles.statistical; style GraphFonts / GraphTitleFont =("Times New Roman",24pt, bold) GraphFootnoteFont =("Times New Roman",24pt, italic) GraphLabelFont =("Times New Roman",24pt) GraphValueFont =("Times New Roman", 24pt) GraphDataFont =("Times New Roman", 24pt) GraphUnicodeFont =(" <MTsans -serif-unicode> ", 24pt) GraphAnnoFont =("Times New Roman", 24pt); end; run; style proc template; path sashelp.tmplmst; list styles; run;

34 / 47 proc template; source Stat.Lifetest.Graphics.ProductLimitSurvival; run;, Template, Stat.Lifetest.Graphics.ProductLimitSurvival

35 / 47 ( ) (134 )

36 / 47 ENTRYTITLE Statement entrytitle "Product -Limit Survival Estimates"; if (EXISTS(SECONDTITLE)) entrytitle SECONDTITLE / textattrs=graphvaluetext; endif;

37 / 47 SCATTERPLOT Statement MARKERATTRS Option if (PLOTCENSORED) scatterplot y=censored x=time / group=stratum index=stratumnum markerattrs=(symbol=plus); endif; MARKERATTRS Option symbol size if (PLOTCENSORED) scatterplot y=censored x=time / group=stratum index=stratumnum markerattrs=(symbol=circlefilled size=8pt); endif;

38 / 47 DISCRETELEGEND Statement TYTLE Option, LOCATION Option, HALIGN Option, VALIGN Option, BORDER Option DiscreteLegend "Survival" / title=groupname location=outside; TYTLE, LOCATION, HALIGN VALIGN, BORDER DiscreteLegend "Survival" / location=inside HALIGN=LEFT VALIGN=BOTTOM border=false; TEMPLATE Procedure proc template;... run;

39 / 47 TEMPLATE Procedure

1.0 Log-rank P = 0.067 Proportion of overall survival 0.8 0.6 0.4 0.2 0.0 No. at risk (1st entry: high, 2nd: middle) 50 36 28 23 15 0 50 42 36 30 25 0 0.0 2.5 5.0 7.5 10.0 12.5 high-risk Months after entry middle-risk TEMPLATE Procedure ( 600 ), ; %km_data() GPLOT Procedure,, 9.1.3 40 / 47

41 / 47 %km_data() ODS Graph / TEMPLATE Procedure http://www.josai.ac.jp/~nagasima/

42 / 47 Example 1 Proportion of overall survival 1.0 0.8 0.6 0.4 0.2 0.0 Log-rank P = 0.030 2:middle-risk HR = 0.611 [0.359, 1.038] 3:low-risk HR = 0.502 [0.289, 0.871] No. at risk (1st entry: high, 2nd: middle, 3rd: low) 50 36 28 23 15 0 50 42 36 30 25 0 50 44 36 32 26 0 0.0 2.5 5.0 7.5 10.0 12.5 high-risk middle-risk low-risk Months after entry

43 / 47 Example 2 Proportion of overall survival 1.0 0.8 0.6 0.4 0.2-2Log(LR) P = 0.029 2:middle-risk HR = 0.611 [0.359, 1.038] 3:low-risk HR = 0.502 [0.289, 0.871] 0.0 0.0 2.5 5.0 7.5 10.0 12.5 high-risk middle-risk low-risk Months after entry

44 / 47 Example 3 Proportion of overall survival 1.0 0.8 0.6 0.4 0.2 0.0 MST 95% C.I. 1: high-risk 7.8 [4.4, 9.0) 2: middle-risk 11.3 [6.7,.) 3: low-risk. [8.2,.) No. at risk (1st entry: high, 2nd: middle, 3rd: low) 50 36 28 23 15 0 50 42 36 30 25 0 50 44 36 32 26 0 0.0 2.5 5.0 7.5 10.0 12.5 high-risk middle-risk low-risk Months after entry

45 / 47 Example 4 1.0 Wilcoxon P = 0.032 Proportion of overall survival 0.8 0.6 0.4 0.2 0.0 No. at risk (1st entry: high, 2nd: middle, 3rd: low) 50 36 28 23 15 0 50 42 36 30 25 0 50 44 36 32 26 0 0.0 2.5 5.0 7.5 10.0 12.5 high-risk middle-risk low-risk Months after entry

46 / 47 Example 5 1.0 Log-rank P = 0.067 Proportion of overall survival 0.8 0.6 0.4 0.2 0.0 No. at risk (1st entry: high, 2nd: middle) 50 36 28 23 15 0 50 42 36 30 25 0 0.0 2.5 5.0 7.5 10.0 12.5 high-risk middle-risk Months after entry

47 / 47 [1] SAS Institute Inc. SAS 9.2 Macro Language: Reference. Cary, NC, USA: SAS Institute Inc., 2009. [2] SAS Institute Inc. SAS/STAT(R) 9.2 User s Guide, Second Edition. Cary, NC, USA: SAS Institute Inc., 2009. [3] SAS Institute Inc. SAS/GRAPH(R) 9.2: Graph Template Language User s Guide, Second Edition. Cary, NC, USA: SAS Institute Inc., 2008. [4] Pocock SJ, Travison TG, Wruck LM. How to interpret figures in reports of clinical trials. BMJ 2008; 336(7654): 1166 1169. [5] Regine WF, Winter KA, Abrams RA, et al. Fluorouracil vs Gemcitabine chemotherapy before and after Fluorouracil-based chemoradiation following resection of pancreatic adenocarcinoma: a randomized controlled trial. JAMA 2008; 299(9): 1019 1026. [6] Reck M, Pawel J, Zatloukal P, et al. Phase III trial of cisplatin plus gemcitabine with either placebo or bevacizumab as first-line therapy for nonsquamous non-small-cell lung cancer: AVAil. J Clin Oncol 2009; 27(8): 1227 1234. [7] Nordlinger B, Sorbye H, Glimelius B, et al. Perioperative chemotherapy with FOLFOX4 and surgery versus surgery alone for resectable liver metastases from colorectal cancer (EORTC Intergroup trial 40983): a randomised controlled trial. Lancet 2008; 371(9617): 1007 1016. [8] Karapetis CS, Khambata-Ford S, Jonker DJ, et al. K-ras mutations and benefit from Cetuximab in advanced colorectal cancer. N Engl J Med 2008; 359(17): 1757-1765.