Review Reports

Round 1

Reviewer 1 Report

Interesting paper - but flawed study design in that it is retrospective and small percentage of patients actually had Oncotype test performed. 

Author Response

Interesting paper - but flawed study design in that it is retrospective and small percentage of patients actually had Oncotype test performed.

Answer) We agree with your point and added the following sentence in the “limitation” of the “Discussion”.

Our study has a major limitation inherent in retrospective analysis from a single institution with a small proportion of patients who received Oncotype DX assay.

As we described in the manuscript, the 21-gene RS assays were tested in 36.2% of the study population because they are not covered by National Health Insurance in Korea. We discussed the use of the 21-gene RS with the patient based on the patient’s clinicopathologic risk factors as well as their personal preferences. The Oncotype DX assay was frequently omitted in patients whose clinicopathologic features were either mostly favorable or poor because their 21-gene RS was strongly expected to be low or high. In contrast, the clinicians generally recommended the application of the 21-gene RS assay to patients with intermediate or mixed clinicopathologic data.

Author Response File: Author Response.pdf

Reviewer 2 Report

Dear Authors,

 

It is an interesting idea, however it needs more work to explain the results:

1) you may describe the 21-genes or at least the pathways.

2) It is not clear what do they mean that the rates of chemotherapies were reduced by the 21 genes-signature.

3) You may add more references in the introduction by describing better the context of their study.

 

 

 

 

Author Response

It is an interesting idea, however it needs more work to explain the results:

1) you may describe the 21-genes or at least the pathways.

Answer) Among the several commercial genomic assays, the 21-gene RS assay (Oncotype DX, Genomic Heal, Redwood City, CA, USA) is the most frequently used in HR-positive, HER2-negative breast cancer because this assay brings value guiding chemotherapy decision based on direct evidence of prediction of chemotherapy benefit, but other genomic assays are prognostic-only. After we delivered the breast tissue to Genomic Health (Redwood City, CA, USA), the company performed the 21-gene RS assay with RNA extraction. The 21-gene RS assay evaluates the expression of 16 tumor-associated genes (ER, PGR, BCL2, SCUBE2, GRB7, HER2, Ki-67, STK15, Survivin, CCNB1, MYBL2, MMP11, CTSL2, GSTM1, CD68, BAG1) and 5 reference genes (ACTB, GAPDH, RPLPO, GUS, TFRC) by using reverse transcriptase-polymerase chain reaction (RT-PCR). Based on the expression levels of 21-genes, an algorithm was designed to compute a 21-gene RS for each sample.    

We revised the “2.3. 21-gene recurrence score assay” in “Materials and Methods” as follows:

The 21-gene RS assay (Oncotype DX, Genomic Health, Redwood City, CA, USA) is based on reverse transcriptase-polymerase chain reaction (RT-PCR) that can be performed on the RNA isolated from formalin-fixed paraffin-embedded (FFPE) tissue. After reviewing the hematoxylin and eosin-stained slides to determine whether sufficient invasive breast cancer was present and whether manual microdissection was indicated, RNA extraction from the unstained sections and 21-gene RS assay was performed by Genomic Health (Redwood City, CA, USA). This assay evaluates the expression of 16 tumor-associated genes (ER, PGR, BCL2, SCUBE2, GRB7, HER2, Ki-67, STK15, Survivin, CCNB1, MYBL2, MMP11, CTSL2, GSTM1, CD68, BAG1) and 5 reference genes (ACTB, GAPDH, RPLPO, GUS, TFRC) by using RT-PCR. Based on the expression levels of 21-genes, an algorithm was designed to compute a 21-gene RS for each sample. Quantitative single-gene scores for ER and PR mRNA expression, determined via an RT-PCR, were also provided within the final assay report by Genomic Health.

 

2) It is not clear what do they mean that the rates of chemotherapies were reduced by the 21 genes-signature.

Answer) As mentioned above, the 21-gene RS assay provides the 21-gene RS, and clinical physicians determined the adjuvant chemotherapy based on the 21-gene RS (Chemotherapy is performed when the 21-gene RS is high, and chemotherapy is not performed when the 21-gene RS is low). Before applying the 21-gene RS, clinical physicians determined the adjuvant chemotherapy based on the clinical risk. However, in the cases without the application of 21-gene RS, there have been concerns that adjuvant chemotherapy was not necessary for the majority of patients, even in those with high clinical risk. In this context, several previous studies reported that the application of 21-gene RS reduced the chemotherapy rate in clinical practice. However, the benefit of the application of 21-gene RS regarding the chemotherapy reduction without adverse survival stratified by clinical risk has not yet been reported. Accordingly, we analyzed the chemotherapy rate and survival according to the application of 21-gene RS in both low and high clinical risk groups.

We found that the application of 21-gene RS significantly reduced the chemotherapy rate in patients at high clinical risk, although there was no difference in patients at low clinical risk. Also, the application of 21-gene RS was an independent factor for chemotherapy rate reduction in patients at high clinical risk among the entire cohort. To address the concerns that significant differences in clinicopathologic factors may influence the chemotherapy rate according to the 21-gene RS in each low and high clinical risk group, we performed the PSM analysis. The multivariable analysis revealed that the application of 21-gene RS was an independent factor for the chemotherapy rate reduction. After the PSM, among the patients at high clinical risk, the chemotherapy rate was lower in patients with application of 21-gene RS than in those without application of 21-gene.

3) You may add more references in the introduction by describing better the context of their study.

Answer) Thank you for your point. We added more references and revised the introduction to clarify the study background.

Reviewer 3 Report

This study could have a strong impact when we talk about the path to personalized medicine. The more I know about my patient, the better I can treat him. Specially in chemotherapy, it is still essential to avoid unnecessary side effects.

The study is well designed, however Tables 2 and 3 are difficult to analyse. The authors could present this data in another way?

The authors present a very complete discussion. However the conclusion is very brief and poor. The conclusion must be improved and must present the main findings of the study.

Author Response

This study could have a strong impact when we talk about the path to personalized medicine. The more I know about my patient, the better I can treat him. Specially in chemotherapy, it is still essential to avoid unnecessary side effects.

The study is well designed, however Tables 2 and 3 are difficult to analyse. The authors could present this data in another way?

Answer) We thought you were pointing out Table 1 and 2, so we revised two Tables to make them as easy to read as possible.

Table 1: We deleted columns 2-4 because we thought that the difference in clinicopathologic factors according to the clinical risk was not necessary in the PSM cohort.

Table 2: Table 2 was too bulky, so only the adjusting OR and p-value for the application of 21-gene RS were left. Covariates used in the multivariable analysis are described in the footnote.

The authors present a very complete discussion. However the conclusion is very brief and poor. The conclusion must be improved and must present the main findings of the study.

Answer) As you pointed out, we modified the “Conclusion” as follow:

In summary, this is the first study to analyze the usefulness of the 21-gene RS according to clinical risk in the subset of patients with tumor size 1–5 cm, N0-1, HR+/HER2- breast cancer, in terms of chemotherapy rate and survival. The application of the 21-gene RS reduced chemotherapy rates, particularly in patients with high clinical risk. In contrast, it did not alter the chemotherapy rate in patients with low clinical risk in the case-matched cohort. Furthermore, the survival outcomes according to implementation of the 21-gene RS did not differ, even in patients at high clinical risk. These data suggest that the 21-gene RS should be considered positively to reduce overtreatment without adverse prognosis in patients with high clinical risk. For the patients with low clinical risk, further studies with long-term follow-up data are warranted to address the role of the 21-gene RS, which could offer chemotherapy for patients with high genomic risk.

Round 2

Reviewer 2 Report

You completed the peer-revision.