Movatterモバイル変換

[0]ホーム
URL:

Umpire 2.0: Simulating AssociatedSurvival

Kevin R. Coombes and Caitlin E. Coombes

Load the packages we need.

library(Umpire)library(survival)

Base Survival

Create a survival model, and generate data.

sm<-SurvivalModel(baseHazard =1/5,# default 1/5 inverse yearsaccrual =5,# default 5 yearsfollowUp =1,# default 1 yearsunits =12,unitName ="months")R<-rand(sm,200)summary(R)
##       LFU          Event        ##  Min.   : 0.00   Mode :logical  ##  1st Qu.:14.75   FALSE:101      ##  Median :28.00   TRUE :99       ##  Mean   :29.25                  ##  3rd Qu.:41.00                  ##  Max.   :70.00

Fit a Kaplan-Meier curve.

baseModel<-survfit(Surv(LFU, Event)~1, R)plot(baseModel)

Base Kaplan-Meier survival curve.

Default method to generate beta coefficients.

Now create several cancer models and generate data.

for (ignorein1:5) {for (jgnorein1:4) {    cm<-CancerModel("survtester",nPossible=20,nPattern=2,SURV =function(n)rnorm(n,0,2),# old default; too largesurvivalModel = sm)    S<-rand(cm,200)    model<-survfit(Surv(LFU, Event)~ CancerSubType, S)print(model)plot(model)  }}
## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  90      9     NA      NA      NA## CancerSubType=2 110    101      8       6       9

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  98      5     NA      NA      NA## CancerSubType=2 102      0     NA      NA      NA

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 101    101      0      NA      NA## CancerSubType=2  99     94      7       5      10

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 103     95      6       5      11## CancerSubType=2  97     97      0       0       0

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  94     81     13      11      19## CancerSubType=2 106     88     12      10      17

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  94     94      1       1       2## CancerSubType=2 106      0     NA      NA      NA

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 101     48     43      30      NA## CancerSubType=2  99     72     22      18      30

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  89      2     NA      NA      NA## CancerSubType=2 111     90     15      11      19

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 100    100      0       0       0## CancerSubType=2 100      0     NA      NA      NA

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 101     12     NA      NA      NA## CancerSubType=2  99     99      0       0       0

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  97      0     NA      NA      NA## CancerSubType=2 103     53     34      27      53

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 102      1     NA      NA      NA## CancerSubType=2  98     93      5       4       6

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 103     98      7       5       9## CancerSubType=2  97      3     NA      NA      NA

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 113     37     NA      64      NA## CancerSubType=2  87     87      1       0       2

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 117      1     NA      NA      NA## CancerSubType=2  83     83      1       1       2

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 103    102      2       2       3## CancerSubType=2  97     12     NA      NA      NA

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 105     18     NA      NA      NA## CancerSubType=2  95     27     NA      57      NA

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  93      4     NA      NA      NA## CancerSubType=2 107     89     15      12      17

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  88     72     11       7      19## CancerSubType=2 112    111      3       2       4

Two group survival curves (original).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 108    108      0       0       0## CancerSubType=2  92     92      0      NA      NA

Two group survival curves (original).

Better method to generate beta coefficients.

Now create several cancer models and generate data.

for (ignorein1:5) {for (jgnorein1:4) {    cm<-CancerModel("survtester",nPossible=20,nPattern=2,SURV =function(n)rnorm(n,0,0.3),survivalModel = sm)    S<-rand(cm,200)    model<-survfit(Surv(LFU, Event)~ CancerSubType, S)print(model)plot(model)  }}
## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  92     40     45      41      NA## CancerSubType=2 108     62     34      27      42

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  89     29     NA      59      NA## CancerSubType=2 111     46     51      39      NA

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  94     35     NA      47      NA## CancerSubType=2 106     39     62      52      NA

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 111     89     17      13      22## CancerSubType=2  89     63     25      15      33

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 102     22     NA      NA      NA## CancerSubType=2  98     78     15      11      22

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  78     54     15      11      22## CancerSubType=2 122     70     28      21      44

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  98     34     NA      41      NA## CancerSubType=2 102     28     NA      57      NA

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  95     43     51      31      NA## CancerSubType=2 105     63     28      22      45

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  98     19     NA      NA      NA## CancerSubType=2 102     29     NA      55      NA

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  97     56     32      27      46## CancerSubType=2 103     37     56      47      NA

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  94     60     26      18      36## CancerSubType=2 106     52     43      23      NA

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 100      5     NA      NA      NA## CancerSubType=2 100     41     54      41      NA

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 106     40     61      48      NA## CancerSubType=2  94     31     68      56      NA

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  96     46     34      27      NA## CancerSubType=2 104     28     NA      56      NA

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 111     87     16      12      21## CancerSubType=2  89     41     40      36      NA

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  96     55     26      20      55## CancerSubType=2 104     53     42      27      56

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 107     65     36      24      46## CancerSubType=2  93     24     63      57      NA

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  98     66     19      17      28## CancerSubType=2 102     51     35      30      49

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  95     73     23      18      29## CancerSubType=2 105     74     21      17      26

Two group survival curves (improved).

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 105     94     11       9      15## CancerSubType=2  95     30     NA      56      NA

Two group survival curves (improved).

Fewer possible hits

for (nPosinc(5,10,15)) {for (jgnorein1:4) {    cm<-CancerModel("survtester",nPossible=nPos,nPattern=2,SURV =function(n)rnorm(n,0,0.3),survivalModel = sm)    S<-rand(cm,200)    model<-survfit(Surv(LFU, Event)~ CancerSubType, S)print(model)plot(model)  }}
## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  90     76     12      11      15## CancerSubType=2 110     97     12       8      19

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 101     18     NA      NA      NA## CancerSubType=2  99     19     NA      NA      NA

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  92     50     34      24      51## CancerSubType=2 108     56     38      28      50

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  95     48     40      29      51## CancerSubType=2 105     48     45      40      NA

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  99     14     NA      NA      NA## CancerSubType=2 101     19     NA      61      NA

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  91     72     15      12      23## CancerSubType=2 109     75     21      18      35

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  96     27     NA      50      NA## CancerSubType=2 104     40     47      38      NA

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  99     47     39      30      NA## CancerSubType=2 101     53     33      30      61

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 100     35     NA      45      NA## CancerSubType=2 100     19     NA      NA      NA

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1 100     48     43      26      NA## CancerSubType=2 100     53     31      23      NA

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  97     27     NA      62      NA## CancerSubType=2 103     44     47      36      NA

## Call: survfit(formula = Surv(LFU, Event) ~ CancerSubType, data = S)## ##                   n events median 0.95LCL 0.95UCL## CancerSubType=1  98     33     NA      53      NA## CancerSubType=2 102     24     NA      62      NA

Appendix

sessionInfo()
## R version 4.4.1 (2024-06-14 ucrt)## Platform: x86_64-w64-mingw32/x64## Running under: Windows 11 x64 (build 26100)## ## Matrix products: default## ## ## locale:## [1] LC_COLLATE=C                          ## [2] LC_CTYPE=English_United States.utf8   ## [3] LC_MONETARY=English_United States.utf8## [4] LC_NUMERIC=C                          ## [5] LC_TIME=English_United States.utf8    ## ## time zone: America/New_York## tzcode source: internal## ## attached base packages:## [1] stats     graphics  grDevices utils     datasets  methods   base     ## ## other attached packages:## [1] survival_3.6-4 Umpire_2.0.11 ## ## loaded via a namespace (and not attached):##  [1] sass_0.4.9         utf8_1.2.4         generics_0.1.3     tidyr_1.3.1       ##  [5] rstatix_0.7.2      lattice_0.22-6     digest_0.6.36      magrittr_2.0.3    ##  [9] evaluate_0.24.0    grid_4.4.1         mvtnorm_1.2-5      fastmap_1.2.0     ## [13] jsonlite_1.8.8     Matrix_1.7-0       backports_1.5.0    mclust_6.1.1      ## [17] purrr_1.0.2        fansi_1.0.6        mc2d_0.2.1         BimodalIndex_1.1.9## [21] scales_1.3.0       jquerylib_0.1.4    abind_1.4-5        cli_3.6.3         ## [25] rlang_1.1.4        munsell_0.5.1      splines_4.4.1      cachem_1.1.0      ## [29] yaml_2.3.10        tools_4.4.1        ggsignif_0.6.4     dplyr_1.1.4       ## [33] colorspace_2.1-1   ggplot2_3.5.1      ggpubr_0.6.0       broom_1.0.6       ## [37] vctrs_0.6.5        R6_2.5.1           lifecycle_1.0.4    car_3.1-2         ## [41] oompaBase_3.2.9    cluster_2.1.6      pkgconfig_2.0.3    pillar_1.9.0      ## [45] bslib_0.8.0        gtable_0.3.5       glue_1.7.0         xfun_0.46         ## [49] tibble_3.2.1       tidyselect_1.2.1   highr_0.11         knitr_1.48        ## [53] htmltools_0.5.8.1  rmarkdown_2.27     carData_3.0-5      compiler_4.4.1
[8]ページ先頭
©2009-2025 Movatter.jp