Outcomes of Intercalary Endoprostheses as a Treatment for Metastases in the Femoral and Humeral Diaphysis

Round 1

Reviewer 1 Report

The authors presented a manuscript describing outcomes of patients with metastatic cancers who were operated due to bone metastases and received intercalary endoprosthesis. The study addresses an interesting question and area that is not usually so widely accepted by many orthopaedic surgeons. A study population is relatively small, showing a low frequency of such procedures (can authors compare the number of patients receiving intercalary endoprosthesis and other orthopedic surgeries associated with metastases in the same center in the same period?). It could be interesting for readers of the journal but in the current form the manuscript has too many major issues that need to be solved to consider the manuscript for publication:

- Overall survival is rather an exploratory analysis in such a case series. Patients with 7 different cancer types were included, each with a very different course of the disease and prognosis. This includes breast cancer or lung cancer which can have very aggressive behavior or renal cell carcinoma which can have a slow course and relatively long survival. Authors should concentrate on aspects related to complications, local relapses, and functional measures. OS should not be presented as the first and main result. The same comment apply to the discussion.

- The study would benefit from a more precise description of ontological status and treatment which patients received before, during, and after the surgery (if any), this is the important factor affecting their survival, Moreover, more detailed information about risk factors is necessary so readers can understand how authors understood "good prognosis" and qualified patients for surgery.

- In the Results, authors write "Local recurrence or death was not recorded until the period of this study", while before they inform about a number of patients who died of disease. The sentence needs clarification

- Please clarify why the number of operations is higher than patients - which patients were operated on more than once and why.

- "Overall 5-year endoprosthesis survival was statistically analyzed by localization (OR 4, 95% CI 0.656-24.36, p = 0.175)." It's not clear if it was the location of the primary tumor or metastases. Was it univariate or multivariate analyses? If univariate, multivariate regression should be performed with other variables that could affect the outcomes

- Patient characteristics should be described together in one table, presenting general descriptive statistics. The current table with each patient characteristic is hard to analyze and can be used as supplementary material

- Statistics are poorly done. Kaplan-Meier curve lacks marks for censored observations. Subgroup comparison is missing information about the type of analyses (univariate or multivariate), and the methods section is missing information about statistics used for this comparison.

- "The MSTS score was statistically significantly (95% CI 23.69–25.48, p = 0.008)" - what is the 95%CI referring to, looks like OR is missing.

- The discussion section lacks information about study limitations.

- Language should be revised and corrected by a native speaker. There are many jargon word.

Author Response

We would like to thank the Reviewer for his or her evaluation of our revised manuscript. Additional comments are addressed below in a point-by-point response. The specific modifications of the manuscript based on the additional comments are highlighted in yellow. Please see the attachment.

Author Response File: Author Response.docx

Reviewer 2 Report

The authors report a single center retrospective case series of 25 patients (27 procedures) that underwent intercalary resection and endoprosthetic reconstruction of the femur or humerus for metastatic bone disease. They conclude that outcomes are excellent, but aseptic loosening is a concern in the humerus (likely due to rotational stresses).

It is important to learn about the outcomes with intercalary resection and endoprosthetic reconstruction, as this procedure is indicated in specific clinical scenarios. The authors present a reasonably large series. However, the description of the methodology, the clarity of the results and the discussion of limitations are all lacking. Please see below for specific points. 

Abstract

• How were the patients identified?

• Please explain the rationale for the time window 

• At what followup visit did the physician complete the MSTS? Was this consistent across all patients?
•  
• What was the minimal followup?

• Please clarify type II and type III failure in the abstract

• Please remove all statistical comparisons. This is a small group of patients and there are many confounding variables. Running underpowered statistical tests to find a significant p-value does not add substance to this retrospective case series. 

Introduction

Line 43: Please clarify ‘for us’. 

• Overall, the Introduction section sets up the rationale for the retrospective review of patient outcomes well. The authors are a bit too proscriptive in describing the fixation options: IMN may be appropriate in some cases with longer life expectancy, and plate and cement fixation in cases with shorter life expectancy. Please revise to that effect.

Methods

• As noted above, please describe in more detail the method of patient identification and source of patient data (medical charts?)

• Were there any exclusion criteria? Were all consecutive patients included?

• Was Ethics approval waived for this review?

• Did any patients receive radiotherapy and/or bone modifying agents such as bisphosphonates? This data is important as these factors can affect implant fixation.

• Please clarify, as noted above, the time point in which the reported MSTS scores were obtained. Latest followup visit?

• How were failures identified? Imaging? Physical exam? Both?

• As noted above, please remove comparative statistical analyses in this small, underpowered, heterogeneous group of patients.

Results

• What is meant by ‘Local recurrence or death was not recorded until the period of this study.’ This contradicts the statement that 44% of patients died of their disease.

• The authors included 2 cases of revision intercalary reconstruction. The risk for failure is increased in revision situations. I suggest excluding those patients, or describing them separately.

• The implant survival statistics are confusing. Did the authors only include patients still alive in the analysis for implant survival.? Was there any loss to followup?

• Did the 7 cases of failure include those that were revisions? Patients that had radiation?

• The difference in outcomes between the femur and tibia should be described descriptively. Statistical analysis is inappropriate for this small dataset. Any p-values that the authors report as statistically significant are likely to be spurious.

• How did the authors determine ‘The complication rate was not associated with the age and gender of patients, section size, type of implant, or histopathology of the primary tumor.’? 

• The clamp-rod interface appears to be a consistent location of fixation failure. It appears that the mode of failure differs between humerus and femur. This can be reported descriptively by the authors.

• MSTS scores: as noted above, please clarify the MSTS score methodology (timing, reported at clinic visit or retrospectively, etc) and remove p-values.

• The authors do no report re-operation rates, which is an objective measure of implant failure.

Discussion

• The Discussion section is clunky and too detailed. I suggest a major streamline, plus discussion of limitations (recall bias, assessment bias, outcomes bias, transfer bias, etc) related to retrospective case series. 

Figures

• Please provide more detail in the figure legends

Author Response

We would like to thank the Reviewer for his or her evaluation of our revised manuscript. Additional comments are addressed below in a point-by-point response. The specific modifications of the manuscript based on the additional comments are highlighted in yellow. Please see the attachment.

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Unfortunately, the clarity and readiness of the manuscript have not improved with the revision. A few major issues need to be still fixed:
- term "overall implant survival" does not make sense since an implant is not a life being and can not have a survival which in medical science is commonly used as survival of patients between specified time points. Please consider another name for the endpoint, e.g. time to implant failure. Definition of endpoint should be provided in methods. It's not clear if the patient's death (with an intact implant) is considered an event or if the observation is censored. Moreover, in abstract it is not clear which overall survival authors are referring to

- all information regarding characteristic of population should be reported in results, not in methods

- Fisher test or Mann Whitney U test are not commonly used for comparison of survival between two groups. According to standards, long-rank test should be used

- black curve on the Kaplan-Meier graph still lacks censoring and is missing in the legend

- information about concomitant systemic therapy has not been provided nor included in the table 1

- limitations are poorly described. Multiple limitations, such as lack of multivariate analyses, concomitant systemic therapy, heterogeneity of patients are not included, whereas these are factors that could affect the outcomes

In the current form, I could recommend publication of the manuscript if the above issues will be appropriately addressed and the study will be published as case series (preferably including the "case series" in the title for clear picture for readers).

Author Response

We would like to thank the Reviewer for his or her evaluation of our revised manuscript. Additional comments are addressed below in a point-by-point response. The specific modifications of the manuscript based on the additional comments are highlighted in yellow.

Comments to final review

Comment: term "overall implant survival" does not make sense since an implant is not a life being and can not have a survival which in medical science is commonly used as survival of patients between specified time points. Please consider another name for the endpoint, e.g. time to implant failure. Definition of endpoint should be provided in methods. It's not clear if the patient's death (with an intact implant) is considered an event or if the observation is censored. Moreover, in abstract it is not clear which overall survival authors are referring to

Answer: We would like to the reviewer for pointing this out. In the abstract, we changed for clarity the term „cumulative survival“ to „cumulative implant survival“. The term „overall implant survival“ is referred to both localisations – humerus, and femur. This term is commonly used in orthopaedic studies as a result we would like to preserve it. The patient's death (with an intact implant) is censored as described in the Kaplan-Meier curve for both localisations. 

Comment: all information regarding characteristic of population should be reported in results, not in methods

and

Comment: Fisher test or Mann Whitney U test are not commonly used for comparison of survival between two groups. According to standards, long-rank test should be used

Answer: We appreciate the Reviewer's reflection. We realise that those changes could be suitable. Although numerous related studies used similar methods and statistical tests. We would surely take this into account in our future papers.

Comment: black curve on the Kaplan-Meier graph still lacks censoring and is missing in the legend

Answer: We would like to thank the Reviewer for observation. The black curve was preserved for clarity. Adding censoring to this summary curve could make the graph complicated.

Comment: information about concomitant systemic therapy has not been provided nor included in the table 1

Answer: We would like to thank the Reviewer for the observation. The scope of this study is to evaluate the implant survival, functional scores, and complications of intercalary endoprostheses. Concomitant therapy does not affect the implant results. For clarity, this issue was added to the limitation as concomitant therapy affects the patient survival – censored observation.

Comment: limitations are poorly described. Multiple limitations, such as lack of multivariate analyses, concomitant systemic therapy, heterogeneity of patients are not included, whereas these are factors that could affect the outcomes

Answer: We would like to thank the Reviewer for observation. We recognize that fact. Limitations were added to the discussion.

Text added:

„This study has several limitations. First stands the design of the study. This is a retrospective study with a small cohort of patients and a mid-term follow-up. Second, due to the limited dataset, our study lacks multivariate analyses. Third, concomitant systemic therapy was not taken into account as does not affect the implant outcome. Considering limitations, we suggest that adequate results were obtained.“