Quartz | Level 8

## Non-inferiority in survival analysis

Dear all,

I am trying to design a study comparing two treatments, an experimental one and a standard of care. The outcome is success / failure, however the status will be checked in various followup times, making the analysis appropriate for survival analysis.

Today I have received some new information, making the primary objective more a non-inferiority one then a superiority. I have found a paper of Jung, Kang, McCall and Blumenstein from 2007, suggesting how to calculate the required sample size. I haven't tried it just yet, but will give it a try manually soon. While thinking about this whole issue, I have realized that even if I will succeed with the sample size calculations, I have no idea how to analyze the data in the future, since I don't know how to perform a non-inferiority test when it comes to the survival analysis.

As far as I understand, if HR denotes the hazard ratio, and HR0 denotes the non-inferiority margin, the null hypothesis is H0: HR>=HR0 vs. H1: HR<HR0

This is as far as I could go. My questions are:

1. Are these the correct hypotheses ?

2. If so, how do I make such an analysis with SAS, is there a special option ? Is there a macro ? In general, is there a guideline of how to analyze non-inferiority studies with SAS ?

Thank you in advance!

5 REPLIES 5
Calcite | Level 5

## Re: Non-inferiority in survival analysis

I have the syntax question regarding non-inferiority survival design using PROC POWER or PROC SEQDESIGN. The non-inferiority margin for HR should be greater than 1 if lower hazard is better. Can SAS provide an example codes?

Obsidian | Level 7

## Re: Non-inferiority in survival analysis

Reviving this post from 2013 and 2016 to see if anyone has any updates/suggestions.

Does anyone know if SAS can calculate a non-inferiority sample size (via a proc or macro)?

Thanks

Calcite | Level 5

## Re: Non-inferiority in survival analysis

Hi All,

could someone please respond to the original question from Blue Nose?

PROC Star

## Re: Non-inferiority in survival analysis

There are a couple of SAS papers that have "non-inferiority" in their titles.  I haven't read them.

At PHUSE 2020:

And at SAS Global Forum 2020:

--------------------------
The hash OUTPUT method will overwrite a SAS data set, but not append. That can be costly. Consider voting for Add a HASH object method which would append a hash object to an existing SAS data set

Would enabling PROC SORT to simultaneously output multiple datasets be useful? Then vote for
Allow PROC SORT to output multiple datasets

--------------------------
Calcite | Level 5

## Re: Non-inferiority in survival analysis

Hi Blue Nose,

Did you find a solution to this answer?

Discussion stats
• 5 replies
• 3186 views
• 1 like
• 5 in conversation