Round 1

Reviewer 1 Report

This paper by Yang and colleagues built a prognostic nomogram to 18 predict overall survival for bladder cancer patients and found that LODDS is a better predictive indicator for bladder cancer patients. It is an interesting paper and I have some questions to be addressed here.

 

1.     Was this model tested on real patients with survival length observed and recorded? Did patients provide any feedback on the model usage?

 

2.     Was the model utilized by a physician and did they provide feedback? Is the model usable and easily understandable to doctors?

 

3.     Are there any further features that might refine the model such that the ROC curve AUC can be closer to 1.0?

 

Author Response

Response to Reviewer 1 Comments

 

This paper by Yang and colleagues built a prognostic nomogram to 18 predict overall survival for bladder cancer patients and found that LODDS is a better predictive indicator for bladder cancer patients. It is an interesting paper and I have some questions to be addressed here.

 

Point 1: Was this model tested on real patients with survival length observed and recorded? Did patients provide any feedback on the model usage?

 

Response 1: We are grateful for the suggestion. To tested this prognostic model on real patients, we carried out external validation using a cohort of 239 real patients in our institution from January 2013 to April 2019. From the results of AUC and Kaplan-Meier curves of risk groups, good discriminating power and predictive accuracy of the model were demonstrated (Figure 7). The prognostic model allows clinicians to identify high-risk patients with poor prognoses. More detailed results of statistical analysis were added in Methods and Results section of the revised manuscript.

 

Point 2: Was the model utilized by a physician and did they provide feedback? Is the model usable and easily understandable to doctors?

 

Response 2: Special thanks to you for your good comments. This prognostic model could be widely utilized by physicians. In clinical practice, clinicians can use our model to postoperatively evaluate the prognoses and risk of patients. Firstly, superimposing all corresponding scores of each variable in a nomogram to calculate the total score of each patient. Secondly, based on the ultimate total score, the survival probability corresponding to the total score can be easily obtained from the nomogram. Finally, patients with a total score greater than 76.28 will be identified as high-risk patients and vice versa as low-risk patients. Prospective utilization of the model is ongoing and we will report the results in another paper.

 

Point 3: Are there any further features that might refine the model such that the ROC curve AUC can be closer to 1.0?

 

Response 3: We are grateful for the suggestion. It is really true as Reviewer suggested that some features may be able to refine the model. This comment is valuable and very helpful for refining and improving our model, as well as the important guiding significance to our ongoing and future research. However, some clinical information cannot be extracted from the SEER database, such as the chemotherapy regimen and the position of positive lymph nodes. Relevant and current researches to refine the model by incorporating more important clinical characteristics and large multi-center trials are ongoing at our institution.

Reviewer 2 Report

1.     The major limitation of this study is that there is no data regarding cancer specific survival. The complete analysis is based on OS. This should be clearly stated in the limitations section. 

2.     How many patients received adjuvant chemotherapy? 

3.     Non-urothelial carcinoma represents a heterogenous group regarding potential aggressiveness. These patients should be excluded from the analysis

4.     English language must be improved

Author Response

Response to Reviewer 2 Comments

Point 1: The major limitation of this study is that there is no data regarding cancer specific survival. The complete analysis is based on OS. This should be clearly stated in the limitations section.

Response 1: Special thanks to you for your good comments. It is really true that there is no data regarding cancer specific survival in our study. As Reviewer suggested that the clear statement of this limitation was added in the Discussion section.

Point 2: How many patients received adjuvant chemotherapy?

Response 2: We are grateful for the question. In this study, 2354 patients received adjuvant chemotherapy. This information was added in the Results section of the revised manuscript.

Point 3: Non-urothelial carcinoma represents a heterogenous group regarding potential aggressiveness. These patients should be excluded from the analysis

Response 3: We are grateful for the suggestion. According with your advice, we excluded 1579 patients with non-urothelial carcinoma from the study. In the revised manuscript, all the statistical analysis was carried out based on the 10938 patients with urothelial carcinoma.

Point 4: English language must be improved

Response 4: We apologize for the language problems in the original manuscript. The language presentation has been improved via English language editing by MDPI. Our manuscript has been checked for correct use of grammar and common technical terms and edited to a level suitable for reporting research in a scholarly journal.

Round 2

Reviewer 2 Report

No further comments