Pricing
Free Trial

Sample Size Determination Example

AREA: Survival Analysis | Oncology

nQuery In Use Background Web Opacity 35

Bayer Liver Disease Research

Title: A Phase III study of sorafenib after transarterial chemoembolisation in Japanese and Korean patients with
unresectable hepatocellular carcinoma

Objective: In Japan and South Korea, transarterial chemoembolisation (TACE) is an important locoregional treatment for patients with unresectable hepatocellular carcinoma (HCC). Sorafenib, a multikinase inhibitor, has been shown effective and safe in patients with advanced HCC. This phase III trial assessed the efficacy and safety of sorafenib in Japanese and Korean patients with unresectable HCC who responded to TACE.

Year: 2011

Source: New England Journal of Cancer

Link: https://www.sciencedirect.com/science/article/pii/S0959804911003248

Clinical Area: Oncology

Sample Size Section in Paper/Protocol:

“Patient sample size was estimated based on TTP. If 30% and 70% of patients achieved CR and non-CR, respectively, in response to TACE, the median TTP for the placebo group in the mixed population would be 5.7 months. Clinically meaningful improvement was defined as median TTP 50% higher in the sorafenib than in the placebo group.

Assuming one formal interim and one final analysis performed using an O’Brien-Fleming-type alpha spending function with a two-sided alpha of 0.05, 318 events would be required to achieve a statistical power of 95%. Accrual of 372 patients (186 in each group) within 18 months would be expected to result in 318 events after 30 months; if 10% of patients were lost to follow-up, 414 patients would have to be randomized to observe 318 events.”


Summary of Necessary Parameter Estimates for Sample Size Calculation:

 Parameter Value
Significance Level 0.05
Expected Placebo Median Survival 5.7 Months
Expected Hazard Ratio 1.5
Accrual Period 18 Months
Maximum Follow Up Time 30 Months
Power 95%
Number of Interim Analyses 1
Alpha Spending Function (from Protocol)
 O'Brien-Fleming

Email me this example in PDF format

Step 1: 
Select the Group Sequential Test of Two Survivals with Accrual (STT15) table from the Study Design Pane.

This can be done using the radio buttons or alternatively, you can use the search bar at the end of the Select Test Design & Goal window.

Worked Example - sorafenib - worked example - img 1 - Select Table stt15

Step 2:
From the Assistants menu select Survival Parameter Converter.

Worked Example - sorafenib - worked example - img 2 - Convert

Convert the median survival time and hazard ratio into ‘’Group Exponential Parameter".

sorafenib - worked example - img 3

Step 3:
Back in table STT15, enter the values for sample size calculation taken from the study design statement.

At the bottom of the table, be sure to change the Number of Looks from the default to “2”. This specifies one interim analysis and one endpoint analysis.

Sample Size Calculator Example- nQuery- Example 23- Img 04- Group Sequential Test of Two Survivals

sorafenib - worked example - before calculation

Step 4:
Now enter the power 95% and click Run.
The sample size and number of events will auto calculate.

sorafenib - worked example - Table STT15


The analysis calculates a sample size of 319 events per group with a power of 95% as per the study design statement. Note that nQuery always rounds up integers thus a small difference to the author's rounded down to number of events of 318. A sample size of 186 per group is same as author's statement.

 

Step 5:
nQuery Advanced also provides graphs. The sample size and number of events will auto calculate. To access plots, go to the menu and select Plot > Your graph

sorafenib - worked example - img plots

Step 6:
Calculate the sample size adjusted for expected 10% dropout

Sample Size Calculator Example- nQuery- Example 23- Img 12- Group Sequential Test of Two Survivals

This gives a final sample size of 414 per group as per study design statement

 

Email me this example in PDF format

Free 14 Day Trial
Demo Icon white
No download required
 
LOCATION
GraphPad Software DBA Statistical Solutions, 225 Franklin St FL 26 Boston, MA, 02110-2853, United States

Copyright © Statsols. All Rights Reserved. Do Not Sell or Share My Personal Information. Privacy Policy .