bigKM: The BIG 1-98 trial dataset for Kaplan-Meier STEPP.

bigKMR Documentation

The BIG 1-98 trial dataset for Kaplan-Meier STEPP.


This data set contains 2,685 patients in the Breast International Group (BIG) 1-98 randomized clinical trial. The BIG 1-98 is a Phase III clinical trial of 8,010 post menopausal women with hormone-receptor-positive early invasive breast cancer who were randomly assigned adjuvant therapy of letrozole or tamoxifen. Patterns of treatment effects for varying levels of the biomarker Ki-67 labeling index, a measure of cell proliferation, were analyzed using STEPP. The STEPP analysis showed that letrozole was more effective than tamoxifen for patients with tumors expressing the highest levels of the Ki-67 labeling index. The two treatment arms are letrozole and tamoxifen.




There are four columns of numeric values: trt (treatment group), time (time to event), event (competing event types), and ki67 (continuous measurement of biomarker Ki-67).


The Breast International Group (BIG) 1-98 Steering Committee and the International Breast Cancer Study Group (IBCSG) are acknowledged for permission to use the data from the BIG 1-98 trial.


Lazar A, Cole B, Bonetti M, Gelber R (2010), "Evaluation of treatment-effect heterogeneity using biomarkers measured on a continuous scale: subpopulation treatment effect pattern plot." J Clin Oncol, 28(29), 4539-44.

Viale G et al (2008), "Prognostic and predictive value of cnetrally reviewed Ki-67 labeling index in postmenopausal women with endocrine-responsive breast cancer: results from Breast International Group Trial 1-98 comparing adjuvant tamoxifen and letrozole." J Clin Oncol, 28(34), 5569-75.



rxgroup <- bigKM$trt
time    <- bigKM$time
evt     <- bigKM$event
cov     <- bigKM$ki67

# analyze using Cumulative Incidence method with
# sliding window size of 150 patients and a maximum of 50 patients in common
swin    <- new("stwin", type="sliding", r1=50, r2=150) # create a sliding window
subp    <- new("stsubpop")                             # create subpopulation object
subp    <- generate(subp, win=swin, covariate=cov) # generate the subpopulations
summary(subp)					   # summary of the subpopulations

# create a stepp model using Kaplan Meier Method to analyze the data
smodel  <- new("stmodelKM", coltrt=rxgroup, trts=c(1,2), survTime=time, censor=evt, timePoint=4)

statKM  <- new("steppes")		  # create a test object based on subpopulation and window
statKM  <- estimate(statKM, subp, smodel) # estimate the subpopulation results
# Warning: In this example, the permutations have been set to 0 to allow the function
# to finish in a short amount of time.  IT IS RECOMMEND TO USE AT LEAST 2500 PERMUTATIONS TO 
statKM  <- test(statKM, nperm=0)       # permutation test with 0 iterations

print(statKM)				  # print the estimates and test statistics
plot(statKM, ncex=0.65, legendy=30, pline=-15.5, color=c("blue","gold"),
        xlabel="Median Ki-67 LI in Subpopulation (% immunoreactivity)",
        ylabel="4-year Disease Free Survival", 
        tlegend=c("Letrozole", "Tamoxifen"), nlas=3)


stepp documentation built on June 22, 2024, 9:24 a.m.

Related to bigKM in stepp...