Tolerability of BRAF and MEK Inhibitors for Metastasized Melanoma after Intra-Class Switch: A Multicenter, Retrospective Study

Round 1

Reviewer 1 Report

In the manuscript entitled “Tolerability of BRAF and MEK inhibitors for metastasized melanoma after intra-class switch: a multicenter, retrospective study”, Salzmann and colleagues aim to evaluate if switching from one targeted therapy regiment to another can improve tolerability.

This is a relevant question with direct implications in the clinical management of metastatic melanoma patients. The conclusions are supported by the presented data, there is no overstatements and the limitations of the study and future complementary studies are clearly stated.

The presentation of the data could be improved by additional Figures that would make it make it more visual as suggested in my comments, but the overall quality of the manuscript is excellent.

Line 139, “56 patients (62%, 8/10) alive at 139 the time of data collection.” What does 8/10 mean?

Table 1 and the description made in the text from the first paragraph of the result section are not describing exactly the same aspects of the cohort. Although there is no contradiction between the two, it could appear confusing to the reader. For example, the text indicates that “10 patients (11%) were treated in an adjuvant setting” while only 7 patients are listed in “Adjuvant Stage III”.

Table 1, please check the numbers for the S100 serum level. If we sum-up the Normal, <2 and >=2 24+16+21=61 and not 68. Likewise, 21/68 is not 41% but 31%.

The distribution of interim treatment between the two BRAFi+MEKi combinations described in the paragraph starting line 153 could be included in Figure 1.

Line 157, please check the sentence “42 patients (45%) received no systemic interim treatment: In this collective…”

The distribution of the patients who discontinued BRAFi+MEKi due to toxicity could be included in a Figure to make it easier to track down the switches from one treatment to another.

In addition to Table 2, it could be interesting to plot the highest-grade AE (or weighted sum of grade 1, 2 and 3 AE) for each combination to the new one. This would be a very visual way to show that AE globally decrease when switching combination therapy. Also, this might reveal that some switches are more beneficial than others.

Table 3 could include a line with the number of patients no longer experiencing DLT after the switch to the new treatment.

Author Response

Dear Reviewer,

thank you very much for your positive, productive feedback on our manuscript entitled “Tolerability of BRAF and MEK inhibitors for metastasized melanoma after intra-class switch: a multicenter, retrospective study”, which helped to enhance the quality of our manuscript. We addressed your comments in the following manner, with changes marked in the revised manuscript:

Reviewer 1:

Comment 1: Line 139, “56 patients (62%, 8/10) alive at 139 the time of data collection.” What does 8/10 mean?

Response 1: 8/10 is the proportion of patients alive among adjuvant patients, referring to the sentence before – this has been clarified (now line 141).

Comment 2: Table 1 and the description made in the text from the first paragraph of the result section are not describing exactly the same aspects of the cohort. Although there is no contradiction between the two, it could appear confusing to the reader. For example, the text indicates that “10 patients (11%) were treated in an adjuvant setting” while only 7 patients are listed in “Adjuvant Stage III”.

Response 2: As mentioned, there is no contradiction, as 3 patients were treated for adjuvant stage IV. This was added in the text to clarify to the reader (line 132).

Comment 3: Table 1, please check the numbers for the S100 serum level. If we sum-up the Normal, <2 and >=2 24+16+21=61 and not 68. Likewise, 21/68 is not 41% but 31%.

Response 3: Thank you very much for thoroughly checking the numbers. Please excuse an error in reporting S100 numbers, which was unnoticed to this point and has been corrected. All other descriptive statistics of table 1 were repeated and double checked to avoid other mistakes in this section.

Comment 4: The distribution of interim treatment between the two BRAFi+MEKi combinations described in the paragraph starting line 153 could be included in Figure 1.

Response 4: An updated version of figure 1 has been included, showing the number of patients with and without interim treatments. Adding the interim treatments into the numbers displayed by arrows would make the graph very complicated, we hope the change was performed as requested.  

Comment 5: Line 157, please check the sentence “42 patients (45%) received no systemic interim treatment: In this collective…”

Unfortunately, we are unsure of the improvement meant to make to the sentence: numbers were double checked, the long sentence was broken up into different parts to facilitate understanding.

Comment 6: The distribution of the patients who discontinued BRAFi+MEKi due to toxicity could be included in a Figure to make it easier to track down the switches from one treatment to another.

Response 6: Figure 2 was added to the manuscript, showing the treatment details of patients who discontinued their first BRAFi+MEKi due to toxicity. For these patients, the respective switch to the second combination with the respective tolerability of the second combination is shown in the figure.

Comment 7: In addition to Table 2, it could be interesting to plot the highest-grade AE (or weighted sum of grade 1, 2 and 3 AE) for each combination to the new one. This would be a very visual way to show that AE globally decrease when switching combination therapy. Also, this might reveal that some switches are more beneficial than others.

Response 7: Thank you for this comment. This issue has been widely addressed by the changes made to table 2 as requested by reviewers 2 and 3, providing more details on the respective highest-grade AEs for each combination. After all, the groups of patients receiving their second BRAFi+MEKi combination are widely made up of a single first BRAFi+MEKi combination as shown in figure 1, making another plot for each switch repetitive (e.g. of 39 patients receiving E+B as their second combination, 35 previously received D+T). The changes to dose-limiting toxicity for each switch are now shown in the added figure 2. We hope the changes address your comment sufficiently. 

Comment 8: Table 3 could include a line with the number of patients no longer experiencing DLT after the switch to the new treatment.

Response 8: This has been added to the table.

Author Response File: Author Response.docx

Reviewer 2 Report

Salzman et al. have conducted a multicenter real-world analysis of "Tolerability of BRAF and MEK inhibitors for metastasized melanoma after intra-class switch". This is a relevant topic in routine clinical practice. Physician who treat metastatic melanoma with BRAF+MEK inhibitors will have patients who need to switch treatment because of intolerable adverse events. 

VC, DT, and EB have different toxicity profiles (VC skin reactions, DT fever etc). I would like to see a table that shows the reason (n and %) for tretment discontinuation in first BRAF+MEKi treatment for VC, DT, and patients separately. I don´t think that it is relevant to pool the number and percentage of AEs of VC, DT, and EB together as done in table 2 (as the number and percentage depend on the combination used). 

Clinically interesting question would have been to compare the PFS of patients who have to switch BRAF+MEK inhibitors because of toxicity to those who continue the same BRAF+MEK inhibitor treatment. If such data is not available for this study, the authors could have compared the PFS of patients with BRAF+MEK switch to PFS data of clinical trials and discuss this topic. 

With these modifications I could recommend this manuscript to be published in Cancers. 

Author Response

Dear Reviewer,

thank you very much for your positive, productive feedback on our manuscript entitled “Tolerability of BRAF and MEK inhibitors for metastasized melanoma after intra-class switch: a multicenter, retrospective study”, which helped to enhance the quality of our manuscript. We addressed your comments in the following manner, with changes marked in the revised manuscript:

Reviewer 2:

Comment 1: VC, DT, and EB have different toxicity profiles (VC skin reactions, DT fever etc). I would like to see a table that shows the reason (n and %) for treatment discontinuation in first BRAF+MEKi treatment for VC, DT, and patients separately. I don´t think that it is relevant to pool the number and percentage of AEs of VC, DT, and EB together as done in table 2 (as the number and percentage depend on the combination used).

Response 1: Table 2 has been thoroughly updated, featuring all AEs of each treatment combination of the first and second combination separately with their respective grade. Pooling of the combinations has been removed. Taking into account the respective comment of Reviewer 3 to include combinations in reporting AEs, table 2 still reports AEs with their respective grade; thus slightly differing from your request to plot the reason for discontinuation. This would be highly repetitive with the reporting of grade 3 AEs. We hope to fulfil the requests of both reviewers with the changes made.

Comment 2: Clinically interesting question would have been to compare the PFS of patients who have to switch BRAF+MEK inhibitors because of toxicity to those who continue the same BRAF+MEK inhibitor treatment. If such data is not available for this study, the authors could have compared the PFS of patients with BRAF+MEK switch to PFS data of clinical trials and discuss this topic.

Response 2: Thank you for this important comment. We did think about optimizing the reporting of efficacy data in our cohort during the conceptualization of the study, but found very inhomogeneous groups to be analyzed and reported: this especially accounts for the patients who received a second combination due to toxicity. These show different durations of their first BRAFi combinations, potential interim treatments, inhomogeneous disease status, and many more potential confounders in a small subgroup. We thus abandoned the idea of recruiting a collective to compare to these patients and focused our work on toxicity. Being of interest to the reader, we tried to form the most relevant collective during this revision: patients with discontinuation due to adverse events, with no progression to their first treatment and no interim treatment. This collective had an exceptionally long PFS both of only the second combination, but also from first initiation of BRAFi+MEKi treatment (added to the results from line 265). The limited validity of these data, indirect comparison with historical cohorts, but also potential explanations of improved efficacy of these patients are now discussed from line 326.

Author Response File: Author Response.docx

Reviewer 3 Report

This work concluded that switching to a different BRAF/MEK inhibitor combination will give a promising improvement in tolerability, but new side effects may occur. The major weakness is that this study did not provide any mechanistic details.

My minor comments:

1. add a citation for "CTCAE".

2. Table 2. Porvide details for "combinations" in addition to merely call them "first" and "second".

3. Table 3. 17 patients showed DLT. 5 same DLT and 13 new DLT. Is this because one of them has both the new and the same DLT? Please clarify.

4. Section 3.3.2. It would be better to provide a table or graph.

5. The author claims that no comparative work is reported between combinations. This should be included to increase the significance of this work. 

Author Response

Dear Reviewer,

thank you very much for your productive feedback on our manuscript entitled “Tolerability of BRAF and MEK inhibitors for metastasized melanoma after intra-class switch: a multicenter, retrospective study”, which helped to enhance the quality of our manuscript. While we are unable to change the lack of mechanistic background due to the study design, we addressed your minor comments in the following manner, with changes marked in the revised manuscript:

Reviewer 3:

Comment 1: Add a citation for "CTCAE".

Response 1: A citation to the National Institute of Health is given in line 101.

Comment 2: Table 2. Provide details for "combinations" in addition to merely call them "first" and "second".

Response 2: In combination with comments of Reviewer 2, table 2 was thoroughly updated: pooling of patients in first and second combination was removed, and all AEs of all combinations for each treatment line have been reported.

Comment 3: Table 3. 17 patients showed DLT. 5 same DLT and 13 new DLT. Is this because one of them has both the new and the same DLT? Please clarify.

Response 3: This is exactly the case and has been clarified in the table description.

Comment 4: Section 3.3.2. It would be better to provide a table or graph.

Response 4: Figure 3 was added, illustrating the number of patients suffering of class-specific side effects and how these changed during the switch to a different combination.

Comment 5: The author claims that no comparative work is reported between combinations. This should be included to increase the significance of this work.

Response 5: This has been added to the introduction as well as the discussion.

Once again, thank you for your comments. We hope these changes are helpful to better describe our results and we look forward to your response.

Author Response File: Author Response.docx

Round 2

Reviewer 3 Report

This is significantly improved.  The pdf file on my end is missing many info on my end. I can only see half of the Table 2.

Author Response

Dear Reviewer,

thank you very much! Please excuse the changes in generating a pdf - I have changed the formatting of table 2 to wide-screen and also uploaded a separate version of the table in the non-published material.

Thank you again for your positive feedback,

on behalf of all authors,

Martin Salzmann