Weighted Trajectory Analysis and Application to Clinical Outcome Assessment

Round 1

Reviewer 1 Report

In this study, authors devised Weighted trajectory analysis (WTA) for clinical outcome assessment. Below are a number of issues that the authors shall address or revise:

1. Authors used only one type of simulation sample in functionality. I wonder whether authors can also apply WTA to other kinds of samples, such as data in TCGA. The same issue also happened in other simulation studies and real world examples.

2. There are also several different analysis methods except for KM analysis. I wonder whether authors can compare WTA to other analysis methods.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

In this paper, inspried by the idea of the Kaplan-Meier estimator and the logrank test, the author adapted the above two methods for charaterizing and comparing the trajectories of ordinal outcomes between multiple groups. Overall, this is an interesting paper. My comments are listed below:

1. In the case of ordinal outcomes, it is not clear to me how the survival function is defined. It looks like the Uj's can be greater than 1. 

2. For section 2.5, it seems like the test between two groups can be easily extended to an ANOVA-type test for multiple group comparison. 

3.   In page 7, below equation (19), could the author explain why the weight  w_{l}is chosen by (−4,−3,−2,−1,0,1,2,3,4)?

1. The authors should make the term "logrank test" consistent across the manuscript (some are "log-rank test").

2. In page 7, all the equations mentioned in the paragraphs seem mislabelled, e.g., in last paragraph, "The asymptotic result in equation (3)" seems to be equation (23). 

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

I am satisfied with the author’s responses to my issues raised in my initial review. I recommend that the revised paper be accepted.